rocksdb/util/rate_limiter.cc
Andrew Kryczka a43481b3d0 Decouple RateLimiter burst size and refill period (#12379)
Summary:
When the rate limiter does not have any waiting requests, the first request to arrive may consume all of the available bandwidth, despite potentially having lower priority than requests that arrive later in the same refill interval. Then, those higher priority requests must wait for a refill. So even in scenarios in which we have an overall bandwidth surplus, the highest priority requests can be sporadically delayed up to a whole refill period.

Alone, this isn't necessarily problematic as the refill period is configurable via `refill_period_us` and can be tuned down as needed until the max sporadic delay is tolerable. However, tuning down `refill_period_us` had a side effect of reducing burst size. Some users require a certain burst size to issue optimal I/O sizes to the underlying storage system.

To satisfy those users, this PR decouples the refill period from the burst size. That way, the max sporadic delay can be limited without impacting I/O sizes issued to the underlying storage system.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/12379

Test Plan:
The goal is to show we can now limit the max sporadic delay without impacting compaction's I/O size.

The benchmark runs compaction with a large I/O size, while user reads simultaneously run at a low rate that does not consume all of the available bandwidth. The max sporadic delay is measured using the P100 of rocksdb.file.read.get.micros. I just used strace to verify the compaction reads follow `rate_limiter_single_burst_bytes`

Setup: `./db_bench -benchmarks=fillrandom,flush -write_buffer_size=67108864 -disable_auto_compactions=true -value_size=256 -num=1048576`

Benchmark: `./db_bench -benchmarks=readrandom -use_existing_db=true -num=1048576 -duration=10 -benchmark_read_rate_limit=4096 -rate_limiter_bytes_per_sec=67108864 -rate_limiter_refill_period_us=$refill_micros -rate_limiter_single_burst_bytes=16777216 -rate_limit_bg_reads=true -rate_limit_user_ops=true -statistics=true -cache_size=0 -stats_level=5 -compaction_readahead_size=16777216 -use_direct_reads=true`

Results:

refill_micros | rocksdb.file.read.get.micros (P100)
-- | --
10000 | 10802
100000 | 100240
1000000 | 922061

For verifying compaction read sizes: `strace -fye pread64 ./db_bench -benchmarks=compact -use_existing_db=true -rate_limiter_bytes_per_sec=67108864 -rate_limiter_refill_period_us=$refill_micros -rate_limiter_single_burst_bytes=16777216 -rate_limit_bg_reads=true -compaction_readahead_size=16777216 -use_direct_reads=true`

Reviewed By: hx235

Differential Revision: D54165675

Pulled By: ajkr

fbshipit-source-id: c5968486316cbfb7ff8e5b7d75d3589883dd1105
2024-02-26 16:55:13 -08:00

392 lines
14 KiB
C++

// 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 <algorithm>
#include "monitoring/statistics_impl.h"
#include "port/port.h"
#include "rocksdb/system_clock.h"
#include "test_util/sync_point.h"
#include "util/aligned_buffer.h"
#include "util/rate_limiter_impl.h"
namespace ROCKSDB_NAMESPACE {
size_t RateLimiter::RequestToken(size_t bytes, size_t alignment,
Env::IOPriority io_priority, Statistics* stats,
RateLimiter::OpType op_type) {
if (io_priority < Env::IO_TOTAL && IsRateLimited(op_type)) {
bytes = std::min(bytes, static_cast<size_t>(GetSingleBurstBytes()));
if (alignment > 0) {
// Here we may actually require more than burst and block
// as we can not write/read less than one page at a time on direct I/O
// thus we do not want to be strictly constrained by burst
bytes = std::max(alignment, TruncateToPageBoundary(alignment, bytes));
}
Request(bytes, io_priority, stats, op_type);
}
return bytes;
}
// Pending request
struct GenericRateLimiter::Req {
explicit Req(int64_t _bytes, port::Mutex* _mu)
: request_bytes(_bytes), bytes(_bytes), cv(_mu) {}
int64_t request_bytes;
int64_t bytes;
port::CondVar cv;
};
GenericRateLimiter::GenericRateLimiter(
int64_t rate_bytes_per_sec, int64_t refill_period_us, int32_t fairness,
RateLimiter::Mode mode, const std::shared_ptr<SystemClock>& clock,
bool auto_tuned, int64_t single_burst_bytes)
: RateLimiter(mode),
refill_period_us_(refill_period_us),
rate_bytes_per_sec_(auto_tuned ? rate_bytes_per_sec / 2
: rate_bytes_per_sec),
refill_bytes_per_period_(
CalculateRefillBytesPerPeriodLocked(rate_bytes_per_sec_)),
raw_single_burst_bytes_(single_burst_bytes),
clock_(clock),
stop_(false),
exit_cv_(&request_mutex_),
requests_to_wait_(0),
available_bytes_(0),
next_refill_us_(NowMicrosMonotonicLocked()),
fairness_(fairness > 100 ? 100 : fairness),
rnd_((uint32_t)time(nullptr)),
wait_until_refill_pending_(false),
auto_tuned_(auto_tuned),
num_drains_(0),
max_bytes_per_sec_(rate_bytes_per_sec),
tuned_time_(NowMicrosMonotonicLocked()) {
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
total_requests_[i] = 0;
total_bytes_through_[i] = 0;
}
}
GenericRateLimiter::~GenericRateLimiter() {
MutexLock g(&request_mutex_);
stop_ = true;
std::deque<Req*>::size_type queues_size_sum = 0;
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
queues_size_sum += queue_[i].size();
}
requests_to_wait_ = static_cast<int32_t>(queues_size_sum);
for (int i = Env::IO_TOTAL - 1; i >= Env::IO_LOW; --i) {
std::deque<Req*> queue = queue_[i];
for (auto& r : queue) {
r->cv.Signal();
}
}
while (requests_to_wait_ > 0) {
exit_cv_.Wait();
}
}
// This API allows user to dynamically change rate limiter's bytes per second.
void GenericRateLimiter::SetBytesPerSecond(int64_t bytes_per_second) {
MutexLock g(&request_mutex_);
SetBytesPerSecondLocked(bytes_per_second);
}
void GenericRateLimiter::SetBytesPerSecondLocked(int64_t bytes_per_second) {
assert(bytes_per_second > 0);
rate_bytes_per_sec_.store(bytes_per_second, std::memory_order_relaxed);
refill_bytes_per_period_.store(
CalculateRefillBytesPerPeriodLocked(bytes_per_second),
std::memory_order_relaxed);
}
Status GenericRateLimiter::SetSingleBurstBytes(int64_t single_burst_bytes) {
if (single_burst_bytes < 0) {
return Status::InvalidArgument(
"`single_burst_bytes` must be greater than or equal to 0");
}
MutexLock g(&request_mutex_);
raw_single_burst_bytes_.store(single_burst_bytes, std::memory_order_relaxed);
return Status::OK();
}
void GenericRateLimiter::Request(int64_t bytes, const Env::IOPriority pri,
Statistics* stats) {
assert(bytes <= GetSingleBurstBytes());
bytes = std::max(static_cast<int64_t>(0), bytes);
TEST_SYNC_POINT("GenericRateLimiter::Request");
TEST_SYNC_POINT_CALLBACK("GenericRateLimiter::Request:1",
&rate_bytes_per_sec_);
MutexLock g(&request_mutex_);
if (auto_tuned_) {
static const int kRefillsPerTune = 100;
std::chrono::microseconds now(NowMicrosMonotonicLocked());
if (now - tuned_time_ >=
kRefillsPerTune * std::chrono::microseconds(refill_period_us_)) {
Status s = TuneLocked();
s.PermitUncheckedError(); //**TODO: What to do on error?
}
}
if (stop_) {
// It is now in the clean-up of ~GenericRateLimiter().
// Therefore any new incoming request will exit from here
// and not get satiesfied.
return;
}
++total_requests_[pri];
if (available_bytes_ > 0) {
int64_t bytes_through = std::min(available_bytes_, bytes);
total_bytes_through_[pri] += bytes_through;
available_bytes_ -= bytes_through;
bytes -= bytes_through;
}
if (bytes == 0) {
return;
}
// Request cannot be satisfied at this moment, enqueue
Req r(bytes, &request_mutex_);
queue_[pri].push_back(&r);
TEST_SYNC_POINT_CALLBACK("GenericRateLimiter::Request:PostEnqueueRequest",
&request_mutex_);
// A thread representing a queued request coordinates with other such threads.
// There are two main duties.
//
// (1) Waiting for the next refill time.
// (2) Refilling the bytes and granting requests.
do {
int64_t time_until_refill_us = next_refill_us_ - NowMicrosMonotonicLocked();
if (time_until_refill_us > 0) {
if (wait_until_refill_pending_) {
// Somebody is performing (1). Trust we'll be woken up when our request
// is granted or we are needed for future duties.
r.cv.Wait();
} else {
// Whichever thread reaches here first performs duty (1) as described
// above.
int64_t wait_until = clock_->NowMicros() + time_until_refill_us;
RecordTick(stats, NUMBER_RATE_LIMITER_DRAINS);
++num_drains_;
wait_until_refill_pending_ = true;
clock_->TimedWait(&r.cv, std::chrono::microseconds(wait_until));
TEST_SYNC_POINT_CALLBACK("GenericRateLimiter::Request:PostTimedWait",
&time_until_refill_us);
wait_until_refill_pending_ = false;
}
} else {
// Whichever thread reaches here first performs duty (2) as described
// above.
RefillBytesAndGrantRequestsLocked();
}
if (r.request_bytes == 0) {
// If there is any remaining requests, make sure there exists at least
// one candidate is awake for future duties by signaling a front request
// of a queue.
for (int i = Env::IO_TOTAL - 1; i >= Env::IO_LOW; --i) {
auto& queue = queue_[i];
if (!queue.empty()) {
queue.front()->cv.Signal();
break;
}
}
}
// Invariant: non-granted request is always in one queue, and granted
// request is always in zero queues.
#ifndef NDEBUG
int num_found = 0;
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
if (std::find(queue_[i].begin(), queue_[i].end(), &r) !=
queue_[i].end()) {
++num_found;
}
}
if (r.request_bytes == 0) {
assert(num_found == 0);
} else {
assert(num_found == 1);
}
#endif // NDEBUG
} while (!stop_ && r.request_bytes > 0);
if (stop_) {
// It is now in the clean-up of ~GenericRateLimiter().
// Therefore any woken-up request will have come out of the loop and then
// exit here. It might or might not have been satisfied.
--requests_to_wait_;
exit_cv_.Signal();
}
}
std::vector<Env::IOPriority>
GenericRateLimiter::GeneratePriorityIterationOrderLocked() {
std::vector<Env::IOPriority> pri_iteration_order(Env::IO_TOTAL /* 4 */);
// We make Env::IO_USER a superior priority by always iterating its queue
// first
pri_iteration_order[0] = Env::IO_USER;
bool high_pri_iterated_after_mid_low_pri = rnd_.OneIn(fairness_);
TEST_SYNC_POINT_CALLBACK(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
"PostRandomOneInFairnessForHighPri",
&high_pri_iterated_after_mid_low_pri);
bool mid_pri_itereated_after_low_pri = rnd_.OneIn(fairness_);
TEST_SYNC_POINT_CALLBACK(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
"PostRandomOneInFairnessForMidPri",
&mid_pri_itereated_after_low_pri);
if (high_pri_iterated_after_mid_low_pri) {
pri_iteration_order[3] = Env::IO_HIGH;
pri_iteration_order[2] =
mid_pri_itereated_after_low_pri ? Env::IO_MID : Env::IO_LOW;
pri_iteration_order[1] =
(pri_iteration_order[2] == Env::IO_MID) ? Env::IO_LOW : Env::IO_MID;
} else {
pri_iteration_order[1] = Env::IO_HIGH;
pri_iteration_order[3] =
mid_pri_itereated_after_low_pri ? Env::IO_MID : Env::IO_LOW;
pri_iteration_order[2] =
(pri_iteration_order[3] == Env::IO_MID) ? Env::IO_LOW : Env::IO_MID;
}
TEST_SYNC_POINT_CALLBACK(
"GenericRateLimiter::GeneratePriorityIterationOrderLocked::"
"PreReturnPriIterationOrder",
&pri_iteration_order);
return pri_iteration_order;
}
void GenericRateLimiter::RefillBytesAndGrantRequestsLocked() {
TEST_SYNC_POINT_CALLBACK(
"GenericRateLimiter::RefillBytesAndGrantRequestsLocked", &request_mutex_);
next_refill_us_ = NowMicrosMonotonicLocked() + refill_period_us_;
// Carry over the left over quota from the last period
auto refill_bytes_per_period =
refill_bytes_per_period_.load(std::memory_order_relaxed);
assert(available_bytes_ == 0);
available_bytes_ = refill_bytes_per_period;
std::vector<Env::IOPriority> pri_iteration_order =
GeneratePriorityIterationOrderLocked();
for (int i = Env::IO_LOW; i < Env::IO_TOTAL; ++i) {
assert(!pri_iteration_order.empty());
Env::IOPriority current_pri = pri_iteration_order[i];
auto* queue = &queue_[current_pri];
while (!queue->empty()) {
auto* next_req = queue->front();
if (available_bytes_ < next_req->request_bytes) {
// Grant partial request_bytes even if request is for more than
// `available_bytes_`, which can happen in a few situations:
//
// - The available bytes were partially consumed by other request(s)
// - The rate was dynamically reduced while requests were already
// enqueued
// - The burst size was explicitly set to be larger than the refill size
next_req->request_bytes -= available_bytes_;
available_bytes_ = 0;
break;
}
available_bytes_ -= next_req->request_bytes;
next_req->request_bytes = 0;
total_bytes_through_[current_pri] += next_req->bytes;
queue->pop_front();
// Quota granted, signal the thread to exit
next_req->cv.Signal();
}
}
}
int64_t GenericRateLimiter::CalculateRefillBytesPerPeriodLocked(
int64_t rate_bytes_per_sec) {
if (std::numeric_limits<int64_t>::max() / rate_bytes_per_sec <
refill_period_us_) {
// Avoid unexpected result in the overflow case. The result now is still
// inaccurate but is a number that is large enough.
return std::numeric_limits<int64_t>::max() / kMicrosecondsPerSecond;
} else {
return rate_bytes_per_sec * refill_period_us_ / kMicrosecondsPerSecond;
}
}
Status GenericRateLimiter::TuneLocked() {
const int kLowWatermarkPct = 50;
const int kHighWatermarkPct = 90;
const int kAdjustFactorPct = 5;
// computed rate limit will be in
// `[max_bytes_per_sec_ / kAllowedRangeFactor, max_bytes_per_sec_]`.
const int kAllowedRangeFactor = 20;
std::chrono::microseconds prev_tuned_time = tuned_time_;
tuned_time_ = std::chrono::microseconds(NowMicrosMonotonicLocked());
int64_t elapsed_intervals = (tuned_time_ - prev_tuned_time +
std::chrono::microseconds(refill_period_us_) -
std::chrono::microseconds(1)) /
std::chrono::microseconds(refill_period_us_);
// We tune every kRefillsPerTune intervals, so the overflow and division-by-
// zero conditions should never happen.
assert(num_drains_ <= std::numeric_limits<int64_t>::max() / 100);
assert(elapsed_intervals > 0);
int64_t drained_pct = num_drains_ * 100 / elapsed_intervals;
int64_t prev_bytes_per_sec = GetBytesPerSecond();
int64_t new_bytes_per_sec;
if (drained_pct == 0) {
new_bytes_per_sec = max_bytes_per_sec_ / kAllowedRangeFactor;
} else if (drained_pct < kLowWatermarkPct) {
// sanitize to prevent overflow
int64_t sanitized_prev_bytes_per_sec =
std::min(prev_bytes_per_sec, std::numeric_limits<int64_t>::max() / 100);
new_bytes_per_sec =
std::max(max_bytes_per_sec_ / kAllowedRangeFactor,
sanitized_prev_bytes_per_sec * 100 / (100 + kAdjustFactorPct));
} else if (drained_pct > kHighWatermarkPct) {
// sanitize to prevent overflow
int64_t sanitized_prev_bytes_per_sec =
std::min(prev_bytes_per_sec, std::numeric_limits<int64_t>::max() /
(100 + kAdjustFactorPct));
new_bytes_per_sec =
std::min(max_bytes_per_sec_,
sanitized_prev_bytes_per_sec * (100 + kAdjustFactorPct) / 100);
} else {
new_bytes_per_sec = prev_bytes_per_sec;
}
if (new_bytes_per_sec != prev_bytes_per_sec) {
SetBytesPerSecondLocked(new_bytes_per_sec);
}
num_drains_ = 0;
return Status::OK();
}
RateLimiter* NewGenericRateLimiter(
int64_t rate_bytes_per_sec, int64_t refill_period_us /* = 100 * 1000 */,
int32_t fairness /* = 10 */,
RateLimiter::Mode mode /* = RateLimiter::Mode::kWritesOnly */,
bool auto_tuned /* = false */, int64_t single_burst_bytes /* = 0 */) {
assert(rate_bytes_per_sec > 0);
assert(refill_period_us > 0);
assert(fairness > 0);
std::unique_ptr<RateLimiter> limiter(new GenericRateLimiter(
rate_bytes_per_sec, refill_period_us, fairness, mode,
SystemClock::Default(), auto_tuned, single_burst_bytes));
return limiter.release();
}
} // namespace ROCKSDB_NAMESPACE