rocksdb/util/rate_limiter_test.cc

594 lines
25 KiB
C++
Raw Normal View History

// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under both the GPLv2 (found in the
// COPYING file in the root directory) and Apache 2.0 License
// (found in the LICENSE.Apache file in the root directory).
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#include <chrono>
#include <cinttypes>
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
#include <cstdint>
#include <limits>
#include "db/db_test_util.h"
#include "port/port.h"
#include "rocksdb/system_clock.h"
Add SystemClock::TimedWait() function (#11753) Summary: Having a synthetic implementation of `TimedWait()` in `SystemClock` will allow us to add `SyncPoint`s while mutex is released, which was previously impossible since the lock was released and reacquired all within `pthread_cond_timedwait()`. Additionally, integrating `TimedWait()` with `MockSystemClock` allows us to cleanup some workarounds in the test code. In this PR I only cleaned up the `GenericRateLimiter` test workaround. This is related to the intended follow-up mentioned in https://github.com/facebook/rocksdb/issues/7101's description. There are a couple differences: (1) This PR does not include removing the particular workaround that initially motivated it. Actually, the `Timer` class uses `InstrumentedCondVar`, so the interface introduced here is inadequate to remove that workaround. On the bright side, the interface introduced in this PR can be changed as needed since it can neither be used nor extended externally, due to using forward-declared `port::CondVar*` in the interface. (2) This PR only makes the change in `SystemClock` not `Env`. Older revisions of this PR included `Env::TimedWait()` and `SpecialEnv::TimedWait()`; however, since they were unused it probably makes sense to defer adding them until when they are needed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/11753 Reviewed By: pdillinger Differential Revision: D48654995 Pulled By: ajkr fbshipit-source-id: 15e19f2454b64d4ec7f50e328691c66ca9911122
2023-08-30 01:39:10 +00:00
#include "test_util/mock_time_env.h"
#include "test_util/sync_point.h"
#include "test_util/testharness.h"
#include "util/random.h"
#include "util/rate_limiter_impl.h"
namespace ROCKSDB_NAMESPACE {
// TODO(yhchiang): the rate will not be accurate when we run test in parallel.
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
class RateLimiterTest : public testing::Test {
protected:
~RateLimiterTest() override {
SyncPoint::GetInstance()->DisableProcessing();
SyncPoint::GetInstance()->ClearAllCallBacks();
}
};
TEST_F(RateLimiterTest, OverflowRate) {
GenericRateLimiter limiter(std::numeric_limits<int64_t>::max(), 1000, 10,
RateLimiter::Mode::kWritesOnly,
SystemClock::Default(), false /* auto_tuned */);
ASSERT_GT(limiter.GetSingleBurstBytes(), 1000000000ll);
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
2015-03-17 21:08:00 +00:00
TEST_F(RateLimiterTest, StartStop) {
std::unique_ptr<RateLimiter> limiter(NewGenericRateLimiter(100, 100, 10));
}
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
TEST_F(RateLimiterTest, GetTotalBytesThrough) {
std::unique_ptr<RateLimiter> limiter(NewGenericRateLimiter(
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
200 /* rate_bytes_per_sec */, 1000 * 1000 /* refill_period_us */,
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
10 /* fairness */));
for (int i = Env::IO_LOW; i <= Env::IO_TOTAL; ++i) {
ASSERT_EQ(limiter->GetTotalBytesThrough(static_cast<Env::IOPriority>(i)),
0);
}
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
std::int64_t request_byte = 200;
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
std::int64_t request_byte_sum = 0;
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
limiter->Request(request_byte, static_cast<Env::IOPriority>(i),
nullptr /* stats */, RateLimiter::OpType::kWrite);
request_byte_sum += request_byte;
}
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
EXPECT_EQ(limiter->GetTotalBytesThrough(static_cast<Env::IOPriority>(i)),
request_byte)
<< "Failed to track total_bytes_through_ correctly when IOPriority = "
<< static_cast<Env::IOPriority>(i);
}
EXPECT_EQ(limiter->GetTotalBytesThrough(Env::IO_TOTAL), request_byte_sum)
<< "Failed to track total_bytes_through_ correctly when IOPriority = "
"Env::IO_TOTAL";
}
TEST_F(RateLimiterTest, GetTotalRequests) {
std::unique_ptr<RateLimiter> limiter(NewGenericRateLimiter(
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
200 /* rate_bytes_per_sec */, 1000 * 1000 /* refill_period_us */,
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
10 /* fairness */));
for (int i = Env::IO_LOW; i <= Env::IO_TOTAL; ++i) {
ASSERT_EQ(limiter->GetTotalRequests(static_cast<Env::IOPriority>(i)), 0);
}
std::int64_t total_requests_sum = 0;
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
limiter->Request(200, static_cast<Env::IOPriority>(i), nullptr /* stats */,
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
RateLimiter::OpType::kWrite);
total_requests_sum += 1;
}
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
EXPECT_EQ(limiter->GetTotalRequests(static_cast<Env::IOPriority>(i)), 1)
<< "Failed to track total_requests_ correctly when IOPriority = "
<< static_cast<Env::IOPriority>(i);
}
EXPECT_EQ(limiter->GetTotalRequests(Env::IO_TOTAL), total_requests_sum)
<< "Failed to track total_requests_ correctly when IOPriority = "
"Env::IO_TOTAL";
}
TEST_F(RateLimiterTest, GetTotalPendingRequests) {
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
std::unique_ptr<RateLimiter> limiter(NewGenericRateLimiter(
200 /* rate_bytes_per_sec */, 1000 * 1000 /* refill_period_us */,
10 /* fairness */));
int64_t total_pending_requests = 0;
for (int i = Env::IO_LOW; i <= Env::IO_TOTAL; ++i) {
ASSERT_OK(limiter->GetTotalPendingRequests(
&total_pending_requests, static_cast<Env::IOPriority>(i)));
ASSERT_EQ(total_pending_requests, 0);
}
// This is a variable for making sure the following callback is called
// and the assertions in it are indeed excuted
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
bool nonzero_pending_requests_verified = false;
SyncPoint::GetInstance()->SetCallBack(
"GenericRateLimiter::Request:PostEnqueueRequest", [&](void* arg) {
port::Mutex* request_mutex = (port::Mutex*)arg;
// We temporarily unlock the mutex so that the following
// GetTotalPendingRequests() can acquire it
request_mutex->Unlock();
for (int i = Env::IO_LOW; i <= Env::IO_TOTAL; ++i) {
EXPECT_OK(limiter->GetTotalPendingRequests(
&total_pending_requests, static_cast<Env::IOPriority>(i)))
<< "Failed to return total pending requests for priority level = "
<< static_cast<Env::IOPriority>(i);
if (i == Env::IO_USER || i == Env::IO_TOTAL) {
EXPECT_EQ(total_pending_requests, 1)
<< "Failed to correctly return total pending requests for "
"priority level = "
<< static_cast<Env::IOPriority>(i);
} else {
EXPECT_EQ(total_pending_requests, 0)
<< "Failed to correctly return total pending requests for "
"priority level = "
<< static_cast<Env::IOPriority>(i);
}
}
// We lock the mutex again so that the request thread can resume running
// with the mutex locked
request_mutex->Lock();
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
nonzero_pending_requests_verified = true;
});
SyncPoint::GetInstance()->EnableProcessing();
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
limiter->Request(200, Env::IO_USER, nullptr /* stats */,
RateLimiter::OpType::kWrite);
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
ASSERT_EQ(nonzero_pending_requests_verified, true);
for (int i = Env::IO_LOW; i <= Env::IO_TOTAL; ++i) {
EXPECT_OK(limiter->GetTotalPendingRequests(&total_pending_requests,
static_cast<Env::IOPriority>(i)))
<< "Failed to return total pending requests for priority level = "
<< static_cast<Env::IOPriority>(i);
EXPECT_EQ(total_pending_requests, 0)
<< "Failed to correctly return total pending requests for priority "
"level = "
<< static_cast<Env::IOPriority>(i);
}
SyncPoint::GetInstance()->DisableProcessing();
SyncPoint::GetInstance()->ClearCallBack(
"GenericRateLimiter::Request:PostEnqueueRequest");
}
TEST_F(RateLimiterTest, Modes) {
for (auto mode : {RateLimiter::Mode::kWritesOnly,
RateLimiter::Mode::kReadsOnly, RateLimiter::Mode::kAllIo}) {
GenericRateLimiter limiter(2000 /* rate_bytes_per_sec */,
1000 * 1000 /* refill_period_us */,
10 /* fairness */, mode, SystemClock::Default(),
false /* auto_tuned */);
limiter.Request(1000 /* bytes */, Env::IO_HIGH, nullptr /* stats */,
RateLimiter::OpType::kRead);
if (mode == RateLimiter::Mode::kWritesOnly) {
ASSERT_EQ(0, limiter.GetTotalBytesThrough(Env::IO_HIGH));
} else {
ASSERT_EQ(1000, limiter.GetTotalBytesThrough(Env::IO_HIGH));
}
limiter.Request(1000 /* bytes */, Env::IO_HIGH, nullptr /* stats */,
RateLimiter::OpType::kWrite);
if (mode == RateLimiter::Mode::kAllIo) {
ASSERT_EQ(2000, limiter.GetTotalBytesThrough(Env::IO_HIGH));
} else {
ASSERT_EQ(1000, limiter.GetTotalBytesThrough(Env::IO_HIGH));
}
}
}
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
TEST_F(RateLimiterTest, GeneratePriorityIterationOrder) {
std::unique_ptr<RateLimiter> limiter(NewGenericRateLimiter(
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
200 /* rate_bytes_per_sec */, 1000 * 1000 /* refill_period_us */,
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
10 /* fairness */));
bool possible_random_one_in_fairness_results_for_high_mid_pri[4][2] = {
{false, false}, {false, true}, {true, false}, {true, true}};
std::vector<Env::IOPriority> possible_priority_iteration_orders[4] = {
{Env::IO_USER, Env::IO_HIGH, Env::IO_MID, Env::IO_LOW},
{Env::IO_USER, Env::IO_HIGH, Env::IO_LOW, Env::IO_MID},
{Env::IO_USER, Env::IO_MID, Env::IO_LOW, Env::IO_HIGH},
{Env::IO_USER, Env::IO_LOW, Env::IO_MID, Env::IO_HIGH}};
for (int i = 0; i < 4; ++i) {
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
// These are variables for making sure the following callbacks are called
// and the assertion in the last callback is indeed excuted
bool high_pri_iterated_after_mid_low_pri_set = false;
bool mid_pri_itereated_after_low_pri_set = false;
bool pri_iteration_order_verified = false;
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
SyncPoint::GetInstance()->SetCallBack(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
"PostRandomOneInFairnessForHighPri",
[&](void* arg) {
bool* high_pri_iterated_after_mid_low_pri = (bool*)arg;
*high_pri_iterated_after_mid_low_pri =
possible_random_one_in_fairness_results_for_high_mid_pri[i][0];
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
high_pri_iterated_after_mid_low_pri_set = true;
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
});
SyncPoint::GetInstance()->SetCallBack(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
"PostRandomOneInFairnessForMidPri",
[&](void* arg) {
bool* mid_pri_itereated_after_low_pri = (bool*)arg;
*mid_pri_itereated_after_low_pri =
possible_random_one_in_fairness_results_for_high_mid_pri[i][1];
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
mid_pri_itereated_after_low_pri_set = true;
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
});
SyncPoint::GetInstance()->SetCallBack(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
"PreReturnPriIterationOrder",
[&](void* arg) {
std::vector<Env::IOPriority>* pri_iteration_order =
(std::vector<Env::IOPriority>*)arg;
EXPECT_EQ(*pri_iteration_order, possible_priority_iteration_orders[i])
<< "Failed to generate priority iteration order correctly when "
"high_pri_iterated_after_mid_low_pri = "
<< possible_random_one_in_fairness_results_for_high_mid_pri[i][0]
<< ", mid_pri_itereated_after_low_pri = "
<< possible_random_one_in_fairness_results_for_high_mid_pri[i][1]
<< std::endl;
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
pri_iteration_order_verified = true;
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
});
SyncPoint::GetInstance()->EnableProcessing();
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
limiter->Request(200 /* request max bytes to drain so that refill and order
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
generation will be triggered every time
GenericRateLimiter::Request() is called */
,
Env::IO_USER, nullptr /* stats */,
RateLimiter::OpType::kWrite);
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
ASSERT_EQ(high_pri_iterated_after_mid_low_pri_set, true);
ASSERT_EQ(mid_pri_itereated_after_low_pri_set, true);
ASSERT_EQ(pri_iteration_order_verified, true);
SyncPoint::GetInstance()->DisableProcessing();
SyncPoint::GetInstance()->ClearCallBack(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
"PreReturnPriIterationOrder");
SyncPoint::GetInstance()->ClearCallBack(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
"PostRandomOneInFairnessForMidPri");
SyncPoint::GetInstance()->ClearCallBack(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
"PostRandomOneInFairnessForHighPri");
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
}
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
2015-03-17 21:08:00 +00:00
TEST_F(RateLimiterTest, Rate) {
auto* env = Env::Default();
struct Arg {
Arg(int32_t _target_rate, int _burst)
: limiter(NewGenericRateLimiter(_target_rate /* rate_bytes_per_sec */,
100 * 1000 /* refill_period_us */,
10 /* fairness */)),
request_size(_target_rate /
10 /* refill period here is 1/10 second */),
burst(_burst) {}
std::unique_ptr<RateLimiter> limiter;
int32_t request_size;
int burst;
};
auto writer = [](void* p) {
const auto& thread_clock = SystemClock::Default();
auto* arg = static_cast<Arg*>(p);
// Test for 2 seconds
auto until = thread_clock->NowMicros() + 2 * 1000000;
Random r((uint32_t)(thread_clock->NowNanos() %
std::numeric_limits<uint32_t>::max()));
while (thread_clock->NowMicros() < until) {
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
for (int i = 0; i < static_cast<int>(r.Skewed(arg->burst * 2) + 1); ++i) {
arg->limiter->Request(r.Uniform(arg->request_size - 1) + 1,
Env::IO_USER, nullptr /* stats */,
RateLimiter::OpType::kWrite);
}
for (int i = 0; i < static_cast<int>(r.Skewed(arg->burst) + 1); ++i) {
arg->limiter->Request(r.Uniform(arg->request_size - 1) + 1,
Env::IO_HIGH, nullptr /* stats */,
RateLimiter::OpType::kWrite);
}
Implement superior user & mid IO priority level in GenericRateLimiter (#8595) Summary: Context: An extra IO_USER priority in rate limiter allows users to optionally charge WAL writes / SST reads to rate limiter at this priority level, which then has higher priority than IO_HIGH and IO_LOW. With an extra IO_USER priority, it allows users to better specify the relative urgency/importance among different requests in rate limiter. As a consequence, IO resource management can better prioritize and limit resource based on user's need. The IO_USER is implemented as superior priority in GenericRateLimiter, in the sense that its request queue will always be iterated first without being constrained to fairness. The reason is that the notion of fairness is only meaningful in helping lower priorities in background IO (i.e, IO_HIGH/MID/LOW) to gain some fair chance to run so that it does not block foreground IO (i.e, the ones that are charged at the level of IO_USER). As we can see, the ultimate goal here is to not blocking foreground IO at IO_USER level, which justifies the superiority of IO_USER. Similar benefits exist for IO_MID priority. - Rewrote the logic of deciding the order of iterating request queues of high/low priorities to include the extra user/mid priority w/o affecting the existing behavior (see PR's [comment](https://github.com/facebook/rocksdb/pull/8595/files#r678749331)) - Included the request queue of user-pri/mid-pri in the code path of next-leader-candidate signaling and GenericRateLimiter's destructor - Included the extra user/mid-pri in bookkeeping data structures: total_bytes_through_ and total_requests_ - Re-written the previous impl of explicitly iterating priorities with a loop from Env::IO_LOW to Env::IO_TOTAL Pull Request resolved: https://github.com/facebook/rocksdb/pull/8595 Test Plan: - passed existing rate_limiter_test.cc - passed added unit tests in rate_limiter_test.cc - run performance test to verify performance with only high/low requests is not affected by this change - Set-up command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom --duration=5 --compression_type=none --num=100000000 --disable_auto_compactions=true --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1))` - Test command: `TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=overwrite --use_existing_db=true --disable_wal=true --duration=30 --compression_type=none --num=100000000 --write_buffer_size=1048576 --writable_file_max_buffer_size=65536 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --level0_slowdown_writes_trigger=$(((1 << 31) - 1)) --level0_stop_writes_trigger=$(((1 << 31) - 1)) --statistics=true --rate_limiter_bytes_per_sec=1048576 --rate_limiter_refill_period_us=1000 --threads=32 |& grep -E '(flush|compact)\.write\.bytes'` - Before (on branch upstream/master): `rocksdb.compact.write.bytes COUNT : 4014162` `rocksdb.flush.write.bytes COUNT : 26715832` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.66 - After (on branch rate_limiter_user_pri): `rocksdb.compact.write.bytes COUNT : 3807822` `rocksdb.flush.write.bytes COUNT : 26098659` rocksdb.flush.write.bytes/rocksdb.compact.write.bytes ~= 6.85 Reviewed By: ajkr Differential Revision: D30577783 Pulled By: hx235 fbshipit-source-id: 0881f2705ffd13ecd331256bde7e8ec874a353f4
2021-08-31 17:59:14 +00:00
for (int i = 0; i < static_cast<int>(r.Skewed(arg->burst / 2 + 1) + 1);
++i) {
arg->limiter->Request(r.Uniform(arg->request_size - 1) + 1, Env::IO_MID,
nullptr /* stats */, RateLimiter::OpType::kWrite);
}
arg->limiter->Request(r.Uniform(arg->request_size - 1) + 1, Env::IO_LOW,
nullptr /* stats */, RateLimiter::OpType::kWrite);
}
};
int samples = 0;
int samples_at_minimum = 0;
for (int i = 1; i <= 16; i *= 2) {
int32_t target = i * 1024 * 10;
Arg arg(target, i / 4 + 1);
int64_t old_total_bytes_through = 0;
for (int iter = 1; iter <= 2; ++iter) {
// second iteration changes the target dynamically
if (iter == 2) {
target *= 2;
arg.limiter->SetBytesPerSecond(target);
}
auto start = env->NowMicros();
for (int t = 0; t < i; ++t) {
env->StartThread(writer, &arg);
}
env->WaitForJoin();
auto elapsed = env->NowMicros() - start;
double rate =
(arg.limiter->GetTotalBytesThrough() - old_total_bytes_through) *
1000000.0 / elapsed;
old_total_bytes_through = arg.limiter->GetTotalBytesThrough();
fprintf(stderr,
"request size [1 - %" PRIi32 "], limit %" PRIi32
" KB/sec, actual rate: %lf KB/sec, elapsed %.2lf seconds\n",
arg.request_size - 1, target / 1024, rate / 1024,
elapsed / 1000000.0);
++samples;
if (rate / target >= 0.80) {
++samples_at_minimum;
}
ASSERT_LE(rate / target, 1.25);
}
}
// This can fail due to slow execution speed, like when using valgrind or in
// heavily loaded CI environments
bool skip_minimum_rate_check =
#if defined(ROCKSDB_VALGRIND_RUN)
true;
#elif defined(OS_MACOSX)
getenv("CIRCLECI") || getenv("GITHUB_ACTIONS");
#else
getenv("SANDCASTLE");
#endif
if (skip_minimum_rate_check) {
fprintf(stderr, "Skipped minimum rate check (%d / %d passed)\n",
samples_at_minimum, samples);
} else {
ASSERT_EQ(samples_at_minimum, samples);
}
}
fix rate limiter to avoid starvation Summary: The current implementation of rate limiter has the possibility to introduce resource starvation when change its limit. This diff aims to fix this problem by consuming request bytes partially. Test Plan: ``` ./rate_limiter_test [==========] Running 4 tests from 1 test case. [----------] Global test environment set-up. [----------] 4 tests from RateLimiterTest [ RUN ] RateLimiterTest.OverflowRate [ OK ] RateLimiterTest.OverflowRate (0 ms) [ RUN ] RateLimiterTest.StartStop [ OK ] RateLimiterTest.StartStop (0 ms) [ RUN ] RateLimiterTest.Rate request size [1 - 1023], limit 10 KB/sec, actual rate: 10.355712 KB/sec, elapsed 2.00 seconds request size [1 - 1023], limit 20 KB/sec, actual rate: 19.136564 KB/sec, elapsed 2.00 seconds request size [1 - 2047], limit 20 KB/sec, actual rate: 20.783976 KB/sec, elapsed 2.10 seconds request size [1 - 2047], limit 40 KB/sec, actual rate: 39.308144 KB/sec, elapsed 2.10 seconds request size [1 - 4095], limit 40 KB/sec, actual rate: 40.318349 KB/sec, elapsed 2.20 seconds request size [1 - 4095], limit 80 KB/sec, actual rate: 79.667396 KB/sec, elapsed 2.20 seconds request size [1 - 8191], limit 80 KB/sec, actual rate: 81.807158 KB/sec, elapsed 2.30 seconds request size [1 - 8191], limit 160 KB/sec, actual rate: 160.659761 KB/sec, elapsed 2.20 seconds request size [1 - 16383], limit 160 KB/sec, actual rate: 160.700990 KB/sec, elapsed 3.00 seconds request size [1 - 16383], limit 320 KB/sec, actual rate: 317.639481 KB/sec, elapsed 2.50 seconds [ OK ] RateLimiterTest.Rate (22618 ms) [ RUN ] RateLimiterTest.LimitChangeTest [COMPLETE] request size 10 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 10 KB, new limit 5KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 10KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 80KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 160KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 320KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 80KB/sec, refill period 1000 ms [ OK ] RateLimiterTest.LimitChangeTest (5002 ms) [----------] 4 tests from RateLimiterTest (27620 ms total) [----------] Global test environment tear-down [==========] 4 tests from 1 test case ran. (27621 ms total) [ PASSED ] 4 tests. ``` Reviewers: sdong, IslamAbdelRahman, yiwu, andrewkr Reviewed By: andrewkr Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D60207
2016-07-01 07:16:29 +00:00
TEST_F(RateLimiterTest, LimitChangeTest) {
// starvation test when limit changes to a smaller value
int64_t refill_period = 1000 * 1000;
auto* env = Env::Default();
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
fix rate limiter to avoid starvation Summary: The current implementation of rate limiter has the possibility to introduce resource starvation when change its limit. This diff aims to fix this problem by consuming request bytes partially. Test Plan: ``` ./rate_limiter_test [==========] Running 4 tests from 1 test case. [----------] Global test environment set-up. [----------] 4 tests from RateLimiterTest [ RUN ] RateLimiterTest.OverflowRate [ OK ] RateLimiterTest.OverflowRate (0 ms) [ RUN ] RateLimiterTest.StartStop [ OK ] RateLimiterTest.StartStop (0 ms) [ RUN ] RateLimiterTest.Rate request size [1 - 1023], limit 10 KB/sec, actual rate: 10.355712 KB/sec, elapsed 2.00 seconds request size [1 - 1023], limit 20 KB/sec, actual rate: 19.136564 KB/sec, elapsed 2.00 seconds request size [1 - 2047], limit 20 KB/sec, actual rate: 20.783976 KB/sec, elapsed 2.10 seconds request size [1 - 2047], limit 40 KB/sec, actual rate: 39.308144 KB/sec, elapsed 2.10 seconds request size [1 - 4095], limit 40 KB/sec, actual rate: 40.318349 KB/sec, elapsed 2.20 seconds request size [1 - 4095], limit 80 KB/sec, actual rate: 79.667396 KB/sec, elapsed 2.20 seconds request size [1 - 8191], limit 80 KB/sec, actual rate: 81.807158 KB/sec, elapsed 2.30 seconds request size [1 - 8191], limit 160 KB/sec, actual rate: 160.659761 KB/sec, elapsed 2.20 seconds request size [1 - 16383], limit 160 KB/sec, actual rate: 160.700990 KB/sec, elapsed 3.00 seconds request size [1 - 16383], limit 320 KB/sec, actual rate: 317.639481 KB/sec, elapsed 2.50 seconds [ OK ] RateLimiterTest.Rate (22618 ms) [ RUN ] RateLimiterTest.LimitChangeTest [COMPLETE] request size 10 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 10 KB, new limit 5KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 10KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 80KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 160KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 320KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 80KB/sec, refill period 1000 ms [ OK ] RateLimiterTest.LimitChangeTest (5002 ms) [----------] 4 tests from RateLimiterTest (27620 ms total) [----------] Global test environment tear-down [==========] 4 tests from 1 test case ran. (27621 ms total) [ PASSED ] 4 tests. ``` Reviewers: sdong, IslamAbdelRahman, yiwu, andrewkr Reviewed By: andrewkr Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D60207
2016-07-01 07:16:29 +00:00
struct Arg {
Arg(int32_t _request_size, Env::IOPriority _pri,
std::shared_ptr<RateLimiter> _limiter)
: request_size(_request_size), pri(_pri), limiter(_limiter) {}
int32_t request_size;
Env::IOPriority pri;
std::shared_ptr<RateLimiter> limiter;
};
auto writer = [](void* p) {
auto* arg = static_cast<Arg*>(p);
arg->limiter->Request(arg->request_size, arg->pri, nullptr /* stats */,
RateLimiter::OpType::kWrite);
fix rate limiter to avoid starvation Summary: The current implementation of rate limiter has the possibility to introduce resource starvation when change its limit. This diff aims to fix this problem by consuming request bytes partially. Test Plan: ``` ./rate_limiter_test [==========] Running 4 tests from 1 test case. [----------] Global test environment set-up. [----------] 4 tests from RateLimiterTest [ RUN ] RateLimiterTest.OverflowRate [ OK ] RateLimiterTest.OverflowRate (0 ms) [ RUN ] RateLimiterTest.StartStop [ OK ] RateLimiterTest.StartStop (0 ms) [ RUN ] RateLimiterTest.Rate request size [1 - 1023], limit 10 KB/sec, actual rate: 10.355712 KB/sec, elapsed 2.00 seconds request size [1 - 1023], limit 20 KB/sec, actual rate: 19.136564 KB/sec, elapsed 2.00 seconds request size [1 - 2047], limit 20 KB/sec, actual rate: 20.783976 KB/sec, elapsed 2.10 seconds request size [1 - 2047], limit 40 KB/sec, actual rate: 39.308144 KB/sec, elapsed 2.10 seconds request size [1 - 4095], limit 40 KB/sec, actual rate: 40.318349 KB/sec, elapsed 2.20 seconds request size [1 - 4095], limit 80 KB/sec, actual rate: 79.667396 KB/sec, elapsed 2.20 seconds request size [1 - 8191], limit 80 KB/sec, actual rate: 81.807158 KB/sec, elapsed 2.30 seconds request size [1 - 8191], limit 160 KB/sec, actual rate: 160.659761 KB/sec, elapsed 2.20 seconds request size [1 - 16383], limit 160 KB/sec, actual rate: 160.700990 KB/sec, elapsed 3.00 seconds request size [1 - 16383], limit 320 KB/sec, actual rate: 317.639481 KB/sec, elapsed 2.50 seconds [ OK ] RateLimiterTest.Rate (22618 ms) [ RUN ] RateLimiterTest.LimitChangeTest [COMPLETE] request size 10 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 10 KB, new limit 5KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 10KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 80KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 160KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 320KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 80KB/sec, refill period 1000 ms [ OK ] RateLimiterTest.LimitChangeTest (5002 ms) [----------] 4 tests from RateLimiterTest (27620 ms total) [----------] Global test environment tear-down [==========] 4 tests from 1 test case ran. (27621 ms total) [ PASSED ] 4 tests. ``` Reviewers: sdong, IslamAbdelRahman, yiwu, andrewkr Reviewed By: andrewkr Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D60207
2016-07-01 07:16:29 +00:00
};
for (uint32_t i = 1; i <= 16; i <<= 1) {
int32_t target = i * 1024 * 10;
// refill per second
for (int iter = 0; iter < 2; iter++) {
std::shared_ptr<RateLimiter> limiter =
std::make_shared<GenericRateLimiter>(
target, refill_period, 10, RateLimiter::Mode::kWritesOnly,
SystemClock::Default(), false /* auto_tuned */);
// After "GenericRateLimiter::Request:1" the mutex is held until the bytes
// are refilled. This test could be improved to change the limit when lock
// is released in `TimedWait()`.
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
fix rate limiter to avoid starvation Summary: The current implementation of rate limiter has the possibility to introduce resource starvation when change its limit. This diff aims to fix this problem by consuming request bytes partially. Test Plan: ``` ./rate_limiter_test [==========] Running 4 tests from 1 test case. [----------] Global test environment set-up. [----------] 4 tests from RateLimiterTest [ RUN ] RateLimiterTest.OverflowRate [ OK ] RateLimiterTest.OverflowRate (0 ms) [ RUN ] RateLimiterTest.StartStop [ OK ] RateLimiterTest.StartStop (0 ms) [ RUN ] RateLimiterTest.Rate request size [1 - 1023], limit 10 KB/sec, actual rate: 10.355712 KB/sec, elapsed 2.00 seconds request size [1 - 1023], limit 20 KB/sec, actual rate: 19.136564 KB/sec, elapsed 2.00 seconds request size [1 - 2047], limit 20 KB/sec, actual rate: 20.783976 KB/sec, elapsed 2.10 seconds request size [1 - 2047], limit 40 KB/sec, actual rate: 39.308144 KB/sec, elapsed 2.10 seconds request size [1 - 4095], limit 40 KB/sec, actual rate: 40.318349 KB/sec, elapsed 2.20 seconds request size [1 - 4095], limit 80 KB/sec, actual rate: 79.667396 KB/sec, elapsed 2.20 seconds request size [1 - 8191], limit 80 KB/sec, actual rate: 81.807158 KB/sec, elapsed 2.30 seconds request size [1 - 8191], limit 160 KB/sec, actual rate: 160.659761 KB/sec, elapsed 2.20 seconds request size [1 - 16383], limit 160 KB/sec, actual rate: 160.700990 KB/sec, elapsed 3.00 seconds request size [1 - 16383], limit 320 KB/sec, actual rate: 317.639481 KB/sec, elapsed 2.50 seconds [ OK ] RateLimiterTest.Rate (22618 ms) [ RUN ] RateLimiterTest.LimitChangeTest [COMPLETE] request size 10 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 10 KB, new limit 5KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 10KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 80KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 160KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 320KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 80KB/sec, refill period 1000 ms [ OK ] RateLimiterTest.LimitChangeTest (5002 ms) [----------] 4 tests from RateLimiterTest (27620 ms total) [----------] Global test environment tear-down [==========] 4 tests from 1 test case ran. (27621 ms total) [ PASSED ] 4 tests. ``` Reviewers: sdong, IslamAbdelRahman, yiwu, andrewkr Reviewed By: andrewkr Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D60207
2016-07-01 07:16:29 +00:00
{{"GenericRateLimiter::Request",
"RateLimiterTest::LimitChangeTest:changeLimitStart"},
{"RateLimiterTest::LimitChangeTest:changeLimitEnd",
"GenericRateLimiter::Request:1"}});
fix rate limiter to avoid starvation Summary: The current implementation of rate limiter has the possibility to introduce resource starvation when change its limit. This diff aims to fix this problem by consuming request bytes partially. Test Plan: ``` ./rate_limiter_test [==========] Running 4 tests from 1 test case. [----------] Global test environment set-up. [----------] 4 tests from RateLimiterTest [ RUN ] RateLimiterTest.OverflowRate [ OK ] RateLimiterTest.OverflowRate (0 ms) [ RUN ] RateLimiterTest.StartStop [ OK ] RateLimiterTest.StartStop (0 ms) [ RUN ] RateLimiterTest.Rate request size [1 - 1023], limit 10 KB/sec, actual rate: 10.355712 KB/sec, elapsed 2.00 seconds request size [1 - 1023], limit 20 KB/sec, actual rate: 19.136564 KB/sec, elapsed 2.00 seconds request size [1 - 2047], limit 20 KB/sec, actual rate: 20.783976 KB/sec, elapsed 2.10 seconds request size [1 - 2047], limit 40 KB/sec, actual rate: 39.308144 KB/sec, elapsed 2.10 seconds request size [1 - 4095], limit 40 KB/sec, actual rate: 40.318349 KB/sec, elapsed 2.20 seconds request size [1 - 4095], limit 80 KB/sec, actual rate: 79.667396 KB/sec, elapsed 2.20 seconds request size [1 - 8191], limit 80 KB/sec, actual rate: 81.807158 KB/sec, elapsed 2.30 seconds request size [1 - 8191], limit 160 KB/sec, actual rate: 160.659761 KB/sec, elapsed 2.20 seconds request size [1 - 16383], limit 160 KB/sec, actual rate: 160.700990 KB/sec, elapsed 3.00 seconds request size [1 - 16383], limit 320 KB/sec, actual rate: 317.639481 KB/sec, elapsed 2.50 seconds [ OK ] RateLimiterTest.Rate (22618 ms) [ RUN ] RateLimiterTest.LimitChangeTest [COMPLETE] request size 10 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 10 KB, new limit 5KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 10KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 80KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 160KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 320KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 80KB/sec, refill period 1000 ms [ OK ] RateLimiterTest.LimitChangeTest (5002 ms) [----------] 4 tests from RateLimiterTest (27620 ms total) [----------] Global test environment tear-down [==========] 4 tests from 1 test case ran. (27621 ms total) [ PASSED ] 4 tests. ``` Reviewers: sdong, IslamAbdelRahman, yiwu, andrewkr Reviewed By: andrewkr Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D60207
2016-07-01 07:16:29 +00:00
Arg arg(target, Env::IO_HIGH, limiter);
// The idea behind is to start a request first, then before it refills,
// update limit to a different value (2X/0.5X). No starvation should
// be guaranteed under any situation
// TODO(lightmark): more test cases are welcome.
env->StartThread(writer, &arg);
int32_t new_limit = (target << 1) >> (iter << 1);
TEST_SYNC_POINT("RateLimiterTest::LimitChangeTest:changeLimitStart");
arg.limiter->SetBytesPerSecond(new_limit);
TEST_SYNC_POINT("RateLimiterTest::LimitChangeTest:changeLimitEnd");
env->WaitForJoin();
fprintf(stderr,
"[COMPLETE] request size %" PRIi32 " KB, new limit %" PRIi32
"KB/sec, refill period %" PRIi64 " ms\n",
target / 1024, new_limit / 1024, refill_period / 1000);
}
}
Improve rate_limiter_test.cc (#8904) Summary: - Fixed a bug in `RateLimiterTest.GeneratePriorityIterationOrder` that the callbacks in this test were not called starting from `i = 1`. Fix by increasing `rate_bytes_per_sec` and requested bytes. - The bug is due to the previous `rate_bytes_per_sec` was set too small, resulting in `refill_bytes_per_period` less than `kMinRefillBytesPerPeriod`. Hence the actual `refill_bytes_per_period` was equal to `kMinRefillBytesPerPeriod` due to the logic [here](https://github.com/facebook/rocksdb/blob/main/util/rate_limiter.cc#L302-L303) and it ended up being greater than the previously set requested bytes. Therefore starting from `i = 1`, `RefillBytesAndGrantRequests()` and `GeneratePriorityIterationOrder` won't be called and the test callbacks was not triggered to execute the assertion. - Added internal flag to assert callbacks are called in `RateLimiterTest.GeneratePriorityIterationOrder` to prevent any future changes defeat the purpose of the test [as suggested](https://github.com/facebook/rocksdb/pull/8890#discussion_r704915134) - Increased `rate_bytes_per_sec` and bytes of each request in `RateLimiterTest.GetTotalBytesThrough`, `RateLimiterTest.GetTotalRequests`, `RateLimiterTest.GetTotalPendingRequests` to trigger the "long path" of execution (i.e, the one trigger RefillBytesAndGrantRequests()) to increase test coverage - This increased the running time of the three tests, see test plan for time difference running locally - Cleared up sync point effects after each test by calling `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearAllCallBacks();` in `~RateLimiterTest()` [as suggested](https://github.com/facebook/rocksdb/pull/8595/files#r697534279) - It's fine to call these two methods even when `EnableProcessing()` or `SetCallBack()` is not called in the test or is already cleaned up. In those cases, calling these two functions in destructor is effectively no-op. - This will allow cleaning up sync point effects of previous test even when the previous test failed in assertion. - Added missing `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in existing tests for completeness - Called `SyncPoint::GetInstance()->DisableProcessing();` and `SyncPoint::GetInstance()->ClearCallBacks(..);` in loop in `RateLimiterTest.GeneratePriorityIterationOrder` for completeness Pull Request resolved: https://github.com/facebook/rocksdb/pull/8904 Test Plan: - Passing existing tests - To verify the 1st change, run `RateLimiterTest.GeneratePriorityIterationOrder` with assertions of callbacks are indeed called under original `rate_bytes_per_sec` and request byte and under updated `rate_bytes_per_sec` and request byte. The former will fail the assertion while the latter succeeds. - Here is the increased test time due to the 3rd change mentioned above in the summary. The relevant 3 tests mentioned in total increase the test time by 6s (~6000/33848 = 17.7% of the original total test time), which IMO is acceptable for better test coverage through running the "long path". - current (run on branch rate_limiter_ut_improve locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (3000 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (3001 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (43349 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (43349 ms total) [ PASSED ] 10 tests. - previous (run on branch main locally) [ RUN ] RateLimiterTest.GetTotalBytesThrough [ OK ] RateLimiterTest.GetTotalBytesThrough (0 ms) [ RUN ] RateLimiterTest.GetTotalRequests [ OK ] RateLimiterTest.GetTotalRequests (0 ms) [ RUN ] RateLimiterTest.GetTotalPendingRequests [ OK ] RateLimiterTest.GetTotalPendingRequests (0 ms) ... [----------] 10 tests from RateLimiterTest (33848 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test case ran. (33848 ms total) [ PASSED ] 10 tests. Reviewed By: ajkr Differential Revision: D30872544 Pulled By: hx235 fbshipit-source-id: ff894f5c1a4bef70e8e407d53b00be45f776b3e4
2021-09-17 15:52:20 +00:00
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
fix rate limiter to avoid starvation Summary: The current implementation of rate limiter has the possibility to introduce resource starvation when change its limit. This diff aims to fix this problem by consuming request bytes partially. Test Plan: ``` ./rate_limiter_test [==========] Running 4 tests from 1 test case. [----------] Global test environment set-up. [----------] 4 tests from RateLimiterTest [ RUN ] RateLimiterTest.OverflowRate [ OK ] RateLimiterTest.OverflowRate (0 ms) [ RUN ] RateLimiterTest.StartStop [ OK ] RateLimiterTest.StartStop (0 ms) [ RUN ] RateLimiterTest.Rate request size [1 - 1023], limit 10 KB/sec, actual rate: 10.355712 KB/sec, elapsed 2.00 seconds request size [1 - 1023], limit 20 KB/sec, actual rate: 19.136564 KB/sec, elapsed 2.00 seconds request size [1 - 2047], limit 20 KB/sec, actual rate: 20.783976 KB/sec, elapsed 2.10 seconds request size [1 - 2047], limit 40 KB/sec, actual rate: 39.308144 KB/sec, elapsed 2.10 seconds request size [1 - 4095], limit 40 KB/sec, actual rate: 40.318349 KB/sec, elapsed 2.20 seconds request size [1 - 4095], limit 80 KB/sec, actual rate: 79.667396 KB/sec, elapsed 2.20 seconds request size [1 - 8191], limit 80 KB/sec, actual rate: 81.807158 KB/sec, elapsed 2.30 seconds request size [1 - 8191], limit 160 KB/sec, actual rate: 160.659761 KB/sec, elapsed 2.20 seconds request size [1 - 16383], limit 160 KB/sec, actual rate: 160.700990 KB/sec, elapsed 3.00 seconds request size [1 - 16383], limit 320 KB/sec, actual rate: 317.639481 KB/sec, elapsed 2.50 seconds [ OK ] RateLimiterTest.Rate (22618 ms) [ RUN ] RateLimiterTest.LimitChangeTest [COMPLETE] request size 10 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 10 KB, new limit 5KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 20 KB, new limit 10KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 80KB/sec, refill period 1000 ms [COMPLETE] request size 40 KB, new limit 20KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 160KB/sec, refill period 1000 ms [COMPLETE] request size 80 KB, new limit 40KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 320KB/sec, refill period 1000 ms [COMPLETE] request size 160 KB, new limit 80KB/sec, refill period 1000 ms [ OK ] RateLimiterTest.LimitChangeTest (5002 ms) [----------] 4 tests from RateLimiterTest (27620 ms total) [----------] Global test environment tear-down [==========] 4 tests from 1 test case ran. (27621 ms total) [ PASSED ] 4 tests. ``` Reviewers: sdong, IslamAbdelRahman, yiwu, andrewkr Reviewed By: andrewkr Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D60207
2016-07-01 07:16:29 +00:00
}
Clean up rate limiter refill logic (#11425) Summary: Context: This pull request update is in response to a comment made on https://github.com/facebook/rocksdb/pull/8596#discussion_r680264932. The current implementation of RefillBytesAndGrantRequestsLocked() drains all available_bytes, but the first request after the last wave of requesting/bytes granting is done is not being handled in the same way. This creates a scenario where if a request for a large amount of bytes is enqueued first, but there are not enough available_bytes to fulfill it, the request is put to sleep until the next refill time. Meanwhile, a later request for a smaller number of bytes comes in and is granted immediately. This behavior is not fair as the earlier request was made first. To address this issue, we have made changes to the code to exhaust the remaining available bytes from the request and queue the remaining. With this change, requests are granted in the order they are received, ensuring that earlier requests are not unfairly delayed by later, smaller requests. The specific scenario described above will no longer occur with this change. Also consolidated `granted` and `request_bytes` as part of the change since `granted` is equivalent to `request_bytes == 0` Pull Request resolved: https://github.com/facebook/rocksdb/pull/11425 Test Plan: Added `AvailableByteSizeExhaustTest` Reviewed By: hx235 Differential Revision: D45570711 Pulled By: jaykorean fbshipit-source-id: a7117ed17bf4b8a7ae0f76124cb41902db1a2592
2023-05-10 17:18:36 +00:00
TEST_F(RateLimiterTest, AvailableByteSizeExhaustTest) {
SpecialEnv special_env(Env::Default(), /*time_elapse_only_sleep*/ true);
const std::chrono::seconds kTimePerRefill(1);
// This test makes sure available_bytes_ get exhausted first before queuing
// any remaining bytes when requested_bytes > available_bytes
const int64_t available_bytes_per_period = 500;
std::shared_ptr<RateLimiter> limiter = std::make_shared<GenericRateLimiter>(
available_bytes_per_period,
std::chrono::microseconds(kTimePerRefill).count(), 10 /* fairness */,
RateLimiter::Mode::kWritesOnly, special_env.GetSystemClock(),
false /* auto_tuned */);
// Step 1. Request 100 and wait for the refill
// so that the remaining available bytes are 400
limiter->Request(100, Env::IO_USER, nullptr /* stats */,
RateLimiter::OpType::kWrite);
special_env.SleepForMicroseconds(
static_cast<int>(std::chrono::microseconds(kTimePerRefill).count()));
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
"GenericRateLimiter::Request:PostEnqueueRequest", [&](void* arg) {
port::Mutex* request_mutex = (port::Mutex*)arg;
request_mutex->Unlock();
// Step 3. Check GetTotalBytesThrough = available_bytes_per_period
// to make sure that the first request (100) and the part of the second
// request (400) made through when the remaining of the second request
// got queued
ASSERT_EQ(available_bytes_per_period,
limiter->GetTotalBytesThrough(Env::IO_USER));
request_mutex->Lock();
});
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
// Step 2. Request 500, which is greater than the remaining available bytes
// (400)
limiter->Request(500, Env::IO_USER, nullptr /* stats */,
RateLimiter::OpType::kWrite);
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearCallBack(
"GenericRateLimiter::Request:PostEnqueueRequest");
}
TEST_F(RateLimiterTest, AutoTuneIncreaseWhenFull) {
const std::chrono::seconds kTimePerRefill(1);
const int kRefillsPerTune = 100; // needs to match util/rate_limiter.cc
Add SystemClock::TimedWait() function (#11753) Summary: Having a synthetic implementation of `TimedWait()` in `SystemClock` will allow us to add `SyncPoint`s while mutex is released, which was previously impossible since the lock was released and reacquired all within `pthread_cond_timedwait()`. Additionally, integrating `TimedWait()` with `MockSystemClock` allows us to cleanup some workarounds in the test code. In this PR I only cleaned up the `GenericRateLimiter` test workaround. This is related to the intended follow-up mentioned in https://github.com/facebook/rocksdb/issues/7101's description. There are a couple differences: (1) This PR does not include removing the particular workaround that initially motivated it. Actually, the `Timer` class uses `InstrumentedCondVar`, so the interface introduced here is inadequate to remove that workaround. On the bright side, the interface introduced in this PR can be changed as needed since it can neither be used nor extended externally, due to using forward-declared `port::CondVar*` in the interface. (2) This PR only makes the change in `SystemClock` not `Env`. Older revisions of this PR included `Env::TimedWait()` and `SpecialEnv::TimedWait()`; however, since they were unused it probably makes sense to defer adding them until when they are needed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/11753 Reviewed By: pdillinger Differential Revision: D48654995 Pulled By: ajkr fbshipit-source-id: 15e19f2454b64d4ec7f50e328691c66ca9911122
2023-08-30 01:39:10 +00:00
auto mock_clock =
std::make_shared<MockSystemClock>(Env::Default()->GetSystemClock());
auto stats = CreateDBStatistics();
std::unique_ptr<RateLimiter> rate_limiter(new GenericRateLimiter(
1000 /* rate_bytes_per_sec */,
std::chrono::microseconds(kTimePerRefill).count(), 10 /* fairness */,
Add SystemClock::TimedWait() function (#11753) Summary: Having a synthetic implementation of `TimedWait()` in `SystemClock` will allow us to add `SyncPoint`s while mutex is released, which was previously impossible since the lock was released and reacquired all within `pthread_cond_timedwait()`. Additionally, integrating `TimedWait()` with `MockSystemClock` allows us to cleanup some workarounds in the test code. In this PR I only cleaned up the `GenericRateLimiter` test workaround. This is related to the intended follow-up mentioned in https://github.com/facebook/rocksdb/issues/7101's description. There are a couple differences: (1) This PR does not include removing the particular workaround that initially motivated it. Actually, the `Timer` class uses `InstrumentedCondVar`, so the interface introduced here is inadequate to remove that workaround. On the bright side, the interface introduced in this PR can be changed as needed since it can neither be used nor extended externally, due to using forward-declared `port::CondVar*` in the interface. (2) This PR only makes the change in `SystemClock` not `Env`. Older revisions of this PR included `Env::TimedWait()` and `SpecialEnv::TimedWait()`; however, since they were unused it probably makes sense to defer adding them until when they are needed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/11753 Reviewed By: pdillinger Differential Revision: D48654995 Pulled By: ajkr fbshipit-source-id: 15e19f2454b64d4ec7f50e328691c66ca9911122
2023-08-30 01:39:10 +00:00
RateLimiter::Mode::kWritesOnly, mock_clock, true /* auto_tuned */));
// verify rate limit increases after a sequence of periods where rate limiter
// is always drained
int64_t orig_bytes_per_sec = rate_limiter->GetSingleBurstBytes();
rate_limiter->Request(orig_bytes_per_sec, Env::IO_HIGH, stats.get(),
RateLimiter::OpType::kWrite);
Add SystemClock::TimedWait() function (#11753) Summary: Having a synthetic implementation of `TimedWait()` in `SystemClock` will allow us to add `SyncPoint`s while mutex is released, which was previously impossible since the lock was released and reacquired all within `pthread_cond_timedwait()`. Additionally, integrating `TimedWait()` with `MockSystemClock` allows us to cleanup some workarounds in the test code. In this PR I only cleaned up the `GenericRateLimiter` test workaround. This is related to the intended follow-up mentioned in https://github.com/facebook/rocksdb/issues/7101's description. There are a couple differences: (1) This PR does not include removing the particular workaround that initially motivated it. Actually, the `Timer` class uses `InstrumentedCondVar`, so the interface introduced here is inadequate to remove that workaround. On the bright side, the interface introduced in this PR can be changed as needed since it can neither be used nor extended externally, due to using forward-declared `port::CondVar*` in the interface. (2) This PR only makes the change in `SystemClock` not `Env`. Older revisions of this PR included `Env::TimedWait()` and `SpecialEnv::TimedWait()`; however, since they were unused it probably makes sense to defer adding them until when they are needed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/11753 Reviewed By: pdillinger Differential Revision: D48654995 Pulled By: ajkr fbshipit-source-id: 15e19f2454b64d4ec7f50e328691c66ca9911122
2023-08-30 01:39:10 +00:00
while (std::chrono::microseconds(mock_clock->NowMicros()) <=
kRefillsPerTune * kTimePerRefill) {
rate_limiter->Request(orig_bytes_per_sec, Env::IO_HIGH, stats.get(),
RateLimiter::OpType::kWrite);
}
int64_t new_bytes_per_sec = rate_limiter->GetSingleBurstBytes();
ASSERT_GT(new_bytes_per_sec, orig_bytes_per_sec);
// decreases after a sequence of periods where rate limiter is not drained
orig_bytes_per_sec = new_bytes_per_sec;
Add SystemClock::TimedWait() function (#11753) Summary: Having a synthetic implementation of `TimedWait()` in `SystemClock` will allow us to add `SyncPoint`s while mutex is released, which was previously impossible since the lock was released and reacquired all within `pthread_cond_timedwait()`. Additionally, integrating `TimedWait()` with `MockSystemClock` allows us to cleanup some workarounds in the test code. In this PR I only cleaned up the `GenericRateLimiter` test workaround. This is related to the intended follow-up mentioned in https://github.com/facebook/rocksdb/issues/7101's description. There are a couple differences: (1) This PR does not include removing the particular workaround that initially motivated it. Actually, the `Timer` class uses `InstrumentedCondVar`, so the interface introduced here is inadequate to remove that workaround. On the bright side, the interface introduced in this PR can be changed as needed since it can neither be used nor extended externally, due to using forward-declared `port::CondVar*` in the interface. (2) This PR only makes the change in `SystemClock` not `Env`. Older revisions of this PR included `Env::TimedWait()` and `SpecialEnv::TimedWait()`; however, since they were unused it probably makes sense to defer adding them until when they are needed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/11753 Reviewed By: pdillinger Differential Revision: D48654995 Pulled By: ajkr fbshipit-source-id: 15e19f2454b64d4ec7f50e328691c66ca9911122
2023-08-30 01:39:10 +00:00
mock_clock->SleepForMicroseconds(static_cast<int>(
kRefillsPerTune * std::chrono::microseconds(kTimePerRefill).count()));
// make a request so tuner can be triggered
rate_limiter->Request(1 /* bytes */, Env::IO_HIGH, stats.get(),
RateLimiter::OpType::kWrite);
new_bytes_per_sec = rate_limiter->GetSingleBurstBytes();
ASSERT_LT(new_bytes_per_sec, orig_bytes_per_sec);
}
TEST_F(RateLimiterTest, WaitHangingBug) {
// At t=0: Threads 0 and 1 request `kBytesPerRefill` bytes at low-pri. One
// will be granted immediately and the other will enter `TimedWait()`.
//
// At t=`kMicrosPerRefill`: Thread 2 requests `kBytesPerRefill` bytes at
// low-pri. Thread 2's request enters the queue. To expose the bug scenario,
// `SyncPoint`s ensure this happens while the lock is temporarily released in
// `TimedWait()`. Before the bug fix, Thread 2's request would then hang in
// `Wait()` interminably.
const int kBytesPerSecond = 100;
const int kMicrosPerSecond = 1000 * 1000;
const int kMicrosPerRefill = kMicrosPerSecond;
const int kBytesPerRefill =
kBytesPerSecond * kMicrosPerRefill / kMicrosPerSecond;
auto mock_clock =
std::make_shared<MockSystemClock>(Env::Default()->GetSystemClock());
std::unique_ptr<RateLimiter> limiter(new GenericRateLimiter(
kBytesPerSecond, kMicrosPerRefill, 10 /* fairness */,
RateLimiter::Mode::kWritesOnly, mock_clock, false /* auto_tuned */));
std::array<std::thread, 3> request_threads;
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->LoadDependency(
{{"RateLimiterTest::WaitHangingBug:InitialRequestsReady",
"MockSystemClock::TimedWait:UnlockedPreSleep"},
{"MockSystemClock::TimedWait:UnlockedPostSleep1",
"RateLimiterTest::WaitHangingBug:TestThreadRequestBegin"},
{"RateLimiterTest::WaitHangingBug:TestThreadRequestEnd",
"MockSystemClock::TimedWait:UnlockedPostSleep2"}});
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
for (int i = 0; i < 2; i++) {
request_threads[i] = std::thread([&]() {
limiter->Request(kBytesPerRefill /* bytes */, Env::IOPriority::IO_LOW,
nullptr /* stats */, RateLimiter::OpType::kWrite);
});
}
while (limiter->GetTotalRequests() < 2) {
}
TEST_SYNC_POINT("RateLimiterTest::WaitHangingBug:InitialRequestsReady");
TEST_SYNC_POINT("RateLimiterTest::WaitHangingBug:TestThreadRequestBegin");
request_threads[2] = std::thread([&]() {
limiter->Request(kBytesPerRefill /* bytes */, Env::IOPriority::IO_LOW,
nullptr /* stats */, RateLimiter::OpType::kWrite);
});
while (limiter->GetTotalRequests() < 3) {
}
TEST_SYNC_POINT("RateLimiterTest::WaitHangingBug:TestThreadRequestEnd");
for (int i = 0; i < 3; i++) {
request_threads[i].join();
}
}
TEST_F(RateLimiterTest, RuntimeSingleBurstBytesChange) {
constexpr int kMicrosecondsPerSecond = 1000000;
const int64_t kRateBytesPerSec = 400;
const int64_t kOldSingleBurstBytes = 100;
const int64_t kOldRefillPeriodUs =
kOldSingleBurstBytes * kMicrosecondsPerSecond / kRateBytesPerSec;
const int64_t kNewSingleBurstBytes = kOldSingleBurstBytes * 2;
SpecialEnv special_env(Env::Default(), /*time_elapse_only_sleep*/ true);
std::unique_ptr<RateLimiter> limiter(new GenericRateLimiter(
kRateBytesPerSec, kOldRefillPeriodUs, 10 /* fairness */,
RateLimiter::Mode::kWritesOnly, special_env.GetSystemClock(),
false /* auto_tuned */));
ASSERT_EQ(kOldSingleBurstBytes, limiter->GetSingleBurstBytes());
ASSERT_TRUE(limiter->SetSingleBurstBytes(0).IsInvalidArgument());
ASSERT_OK(limiter->SetSingleBurstBytes(kNewSingleBurstBytes));
ASSERT_EQ(kNewSingleBurstBytes, limiter->GetSingleBurstBytes());
// If the updated single burst bytes is not reflected in the bytes
// granting process, this request will hang forever.
limiter->Request(limiter->GetSingleBurstBytes() /* bytes */,
Env::IOPriority::IO_USER, nullptr /* stats */,
RateLimiter::OpType::kWrite);
}
} // namespace ROCKSDB_NAMESPACE
int main(int argc, char** argv) {
ROCKSDB_NAMESPACE::port::InstallStackTraceHandler();
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
2015-03-17 21:08:00 +00:00
::testing::InitGoogleTest(&argc, argv);
return RUN_ALL_TESTS();
}