Summary:
[Apache Kvrocks](https://github.com/apache/kvrocks) is an open-source distributed key-value NoSQL database built on top of RocksDB. It serves as a cost-saving and capacity-increasing alternative drop-in replacement for Redis
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11779
Reviewed By: ajkr
Differential Revision: D48872257
Pulled By: cbi42
fbshipit-source-id: 507f67d69b826607a1464a22ec7c60abe11c5124
Summary:
This change add an experimental next-generation HyperClockCache (HCC) with automatic sizing of the underlying hash table. Both the existing version (stable) and the new version (experimental for now) of HCC are available depending on whether an estimated average entry charge is provided in HyperClockCacheOptions.
Internally, we call the two implementations AutoHyperClockCache (new) and FixedHyperClockCache (existing). The performance characteristics and much of the underlying logic are similar enough that AutoHCC is likely to make FixedHCC obsolete, and so it's best considered an evolution of the same technology or solution rather than an alternative. More specifically, both implementations share essentially the same logic for managing the state of individual entries in the cache, including metadata for reference counting and counting clocks for eviction. This metadata, which I like to call the "low-level HCC protocol," includes a read-write lock on entries, but relaxed consistency requirements on the cache (e.g. allowing rare duplication) means high-level cache operations never wait for these low-level per-entry locks. FixedHCC is fully wait-free.
AutoHCC is different in how entries are indexed into an efficient hash table. AutoHCC is "essentially wait-free" as there is no pattern of typical high-level operations on a large cache that can lead to one thread waiting on another to complete some work, though it can happen in some unusual/unlucky cases, or atypical uses such as erasing specific cache keys. Table growth and entry reclamation is more complex in AutoHCC compared to FixedHCC, so uses some localized locking to manage that. AutoHCC uses linear hashing to grow the table as needed, with low latency and to a precise size. AutoHCC depends on anonymous mmap support from the OS (currently verified working on Linux, MacOS, and Windows) to allow the array underlying a hash table to grow in place without wasting resident memory on space reserved but unused. AutoHCC uses a form of chaining while FixedHCC uses open addressing and double hashing.
More specifics:
* In developing this PR, a rare availability bug (minor) was noticed in the existing HCC implementation of Release()+erase_if_last_ref, which is now inherited into AutoHCC. Fixing this without a performance regression will not be simple, so is left for follow-up work.
* Some existing unit tests required adjustment of operational parameters or conditions to work with the new behaviors of AutoHCC. A number of bugs were found and fixed in the validation process, including getting unit tests in good working order.
* Added an option to cache_bench, `-degenerate_hash_bits` for correctness stress testing described below. For this, the tool uses the reverse-engineered hash function for HCC to generate keys in which the specified number of hash bits, in critical positions, have a fixed value. Essentially each degenerate hash bit will half the number of chain heads utilized and double the average chain length.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11738
Test Plan:
unit tests updated, and already added to db crash test. Also
## Correctness
The code includes generous assertions to check for unexpected states, especially at destruction time, so should be able to detect critical concurrency bugs. Less serious "availability bugs" in which cache data is hidden or cleanly lost are more difficult to detect, but also less scary for data correctness (as long as performance is good and the design is sound).
In average operation, the structure is extremely low stress and low contention (see next section) so stressing the corner case logic requires artificially stressing the operating conditions. First, we keep the structure small to increase the number of threads hitting the same chain or entry, and just one cache shard. Second, we artificially degrade the hashing so that chains are much longer than typical, using the new `-degenerate_hash_bits` option to cache_bench. Third, we re-create the structure from scratch frequently in order to exercise the Grow logic repeatedly and to get the benefit of the consistency checks in the structure's destructor in debug builds. For cache_bench this also means disabling the single-threaded "populate cache" step (normally used for steady state performance testing). And of course use many more threads than cores to have many preemptions.
An effective test for working out bugs was this (using debug build of course):
```
while ./cache_bench -cache_type=auto_hyper_clock_cache -histograms=0 -cache_size=8000000 -threads=100 -populate_cache=0 -ops_per_thread=10000 -degenerate_hash_bits=6 -num_shard_bits=0; do :; done
```
Or even smaller cases. This setup has around 27 utilized chains, with around 35 entries each, and yield-waits more than 1 million times per second (very high contention; see next section). I have let this run for hours searching for any lingering issues.
I've also run cache_bench under ASAN, UBSAN, and TSAN.
## Essentially wait free
There is a counter for number of yield() calls when one thread is waiting on another. When we pre-populate the structure in a single thread,
```
./cache_bench -cache_type=auto_hyper_clock_cache -histograms=0 -populate_cache=1 -ops_per_thread=200000 2>&1 | grep Yield
```
We see something on the order of 1 yield call per second across 16 threads, even when we load the system other other jobs (parallel compilation). With -populate_cache=0, there are more yield opportunities with parallel table growth. On an otherwise unloaded system, we still see very small (single digit) yield counts, with a chance of getting into the thousands, and getting into 10s of thousands per second during table growth phase if the system is loaded with other jobs. However, I am not worried about this if performance is still good (see next section).
## Overall performance
Although cache_bench initially suggested performance very close to FixedHCC, there was a very noticeable performance hit under a db_bench setup like used in validating https://github.com/facebook/rocksdb/issues/10626. Much of the difference has been reduced by optimizing Lookup with a "naive" pass that will almost always find entries quickly, and only falling back to the careful Lookup algorithm when not found in the first pass.
Setups (chosen to be sensitive to block cache performance), and compiled with USE_CLANG=1 JEMALLOC=1 PORTABLE=0 DEBUG_LEVEL=0:
```
TEST_TMPDIR=/dev/shm base/db_bench -benchmarks=fillrandom -num=30000000 -disable_wal=1 -bloom_bits=16
```
### No regression on FixedHCC
Running before & after builds at the same time on a 48 core machine.
```
TEST_TMPDIR=/dev/shm /usr/bin/time ./db_bench -benchmarks=readrandom[-X10],block_cache_entry_stats,cache_report_problems -readonly -num=30000000 -bloom_bits=16 -cache_index_and_filter_blocks=1 -cache_size=610000000 -duration 20 -threads=24 -cache_type=fixed_hyper_clock_cache -seed=1234
```
Before:
readrandom [AVG 10 runs] : 847234 (± 8150) ops/sec; 59.2 (± 0.6) MB/sec
703MB max RSS
After:
readrandom [AVG 10 runs] : 851021 (± 7929) ops/sec; 59.5 (± 0.6) MB/sec
706MB max RSS
Probably no material difference.
### Single-threaded performance
Using `[-X2]` and `-threads=1` and `-duration=30`, running all three at the same time:
lru_cache: 55100 ops/sec, then 55862 ops/sec (627MB max RSS)
fixed_hyper_clock_cache: 60496 ops/sec, then 61231 ops/sec (626MB max RSS)
auto_hyper_clock_cache: 47560 ops/sec, then 56081 ops/sec (626MB max RSS)
So AutoHCC has more ramp-up cost in the first pass as the cache grows to the appropriate size. (In single-threaded operation, the parallelizability and per-op low latency of table growth is overall slower.) However, once up to size, its performance is comparable to LRUCache. FixedHCC's lean operations still win overall when a good estimate is available.
If we look at HCC table stats, we can see that this configuration is not favorable to AutoHCC (and I have verified that other memory sizes do not yield substantially different results, until shards are under-sized for the full filters):
FixedHCC:
Slot occupancy stats: Overall 47% (124991/262144), Min/Max/Window = 28%/64%/500, MaxRun{Pos/Neg} = 17/22
AutoHCC:
Slot occupancy stats: Overall 59% (125781/209682), Min/Max/Window = 43%/82%/500, MaxRun{Pos/Neg} = 76/16
Head occupancy stats: Overall 43% (92259/209682), Min/Max/Window = 24%/74%/500, MaxRun{Pos/Neg} = 19/26
Entries at home count: 53350
FixedHCC configuration is relatively good for speed, and not ideal for space utilization. As is typical, AutoHCC has tighter control on metadata usage (209682 x 64 bytes rather than 262144 x 64 bytes), and the higher load factor is slightly worse for speed. LRUCache also has more metadata usage, at 199680 x 96 bytes of tracked metadata (plus roughly another 10% of that untracked in the head pointers), and that metadata is subject to fragmentation.
### Parallel performance, high hit rate
Now using `[-X10]` and `-threads=10`, all three at the same time
lru_cache: [AVG 10 runs] : 263629 (± 1425) ops/sec; 18.4 (± 0.1) MB/sec
655MB max RSS, 97.1% cache hit rate
fixed_hyper_clock_cache: [AVG 10 runs] : 479590 (± 8114) ops/sec; 33.5 (± 0.6) MB/sec
651MB max RSS, 97.1% cache hit rate
auto_hyper_clock_cache: [AVG 10 runs] : 418687 (± 5915) ops/sec; 29.3 (± 0.4) MB/sec
657MB max RSS, 97.1% cache hit rate
Even with just 10-way parallelism for each cache (though 30+/48 cores busy overall), LRUCache is already showing performance degradation, while AutoHCC is in the neighborhood of FixedHCC. And that brings us to the question of how AutoHCC holds up under extreme parallelism, so now independent runs with `-threads=100` (overloading 48 cores).
lru_cache: 438613 ops/sec, 827MB max RSS
fixed_hyper_clock_cache: 1651310 ops/sec, 812MB max RSS
auto_hyper_clock_cache: 1505875 ops/sec, 821MB max RSS (Yield count: 1089 over 30s)
Clearly, AutoHCC holds up extremely well under extreme parallelism, even closing some of the modest performance gap with FixedHCC.
### Parallel performance, low hit rate
To get down to roughly 50% cache hit rate, we use `-cache_index_and_filter_blocks=0 -cache_size=1650000000` with `-threads=10`. Here the extra cost of running counting clock eviction, especially on the chains of AutoHCC, are evident, especially with the lower contention of cache_index_and_filter_blocks=0:
lru_cache: 725231 ops/sec, 1770MB max RSS, 51.3% hit rate
fixed_hyper_clock_cache: 638620 ops/sec, 1765MB max RSS, 50.2% hit rate
auto_hyper_clock_cache: 541018 ops/sec, 1777MB max RSS, 50.8% hit rate
Reviewed By: jowlyzhang
Differential Revision: D48784755
Pulled By: pdillinger
fbshipit-source-id: e79813dc087474ac427637dd282a14fa3011a6e4
Summary:
This should only affect iterator when
- user uses DeleteRange(),
- An iterator from level L has a non-ok status (such non-ok status may not be caught before the bug fix in https://github.com/facebook/rocksdb/pull/11783), and
- A range tombstone covers a key from level > L and triggers a reseek sets the status_ to OK in SeekImpl()/SeekPrevImpl() e.g. bd6a8340c3/table/merging_iterator.cc (L801)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11786
Differential Revision: D48908830
Pulled By: cbi42
fbshipit-source-id: eb564be375af4e33dc27542eff753260186e6d5d
Summary:
This happens in (Compaction)MergingIterator layer, and can cause data loss during compaction or read/scan return incorrect result
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11782
Reviewed By: ajkr
Differential Revision: D48880575
Pulled By: cbi42
fbshipit-source-id: 2294ad284a6d653d3674bebe55380f12ee4b645b
Summary:
**Context/Summary:**
After https://github.com/facebook/rocksdb/pull/11631, we rely on `compaction_readahead_size` for how much to read ahead for compaction read under non-direct IO case. https://github.com/facebook/rocksdb/pull/11658 therefore also sanitized 0 `compaction_readahead_size` to 2MB under non-direct IO, which is consistent with the existing sanitization with direct IO.
However, this makes disabling compaction readahead impossible as well as add one more scenario to the inconsistent effects between `Options.compaction_readahead_size=0` during DB open and `SetDBOptions("compaction_readahead_size", "0")` .
- `SetDBOptions("compaction_readahead_size", "0")` will disable compaction readahead as its logic never goes through sanitization above while `Options.compaction_readahead_size=0` will go through sanitization.
Therefore we decided to do this PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11762
Test Plan: Modified existing UTs to cover this PR
Reviewed By: ajkr
Differential Revision: D48759560
Pulled By: hx235
fbshipit-source-id: b3f85e58bda362a6fa1dc26bd8a87aa0e171af79
Summary:
For a SST file that uses user-defined timestamp aware comparators, if a lower or upper bound is set, sst_dump tool doesn't handle it well. This PR adds support for that. While working on this `MaybeAddTimestampsToRange` is moved to the udt_util.h file to be shared.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11757
Test Plan:
make all check
for changes in db_impl.cc and db_impl_compaction_flush.cc
for changes in sst_file_dumper.cc, I manually tested this change handles specifying bounds for UDT use cases. It probably should have a unit test file eventually.
Reviewed By: ltamasi
Differential Revision: D48668048
Pulled By: jowlyzhang
fbshipit-source-id: 1560465f40e44668d6d82a7439fe9012be0e74a8
Summary:
wide_columns can now be pretty-printed in the following commands
- `./ldb dump_wal`
- `./ldb dump`
- `./ldb idump`
- `./ldb dump_live_files`
- `./ldb scan`
- `./sst_dump --command=scan`
There are opportunities to refactor to reduce some nearly identical code. This PR is initial change to add wide column support in `ldb` and `sst_dump` tool. More PRs to come for the refactor.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11754
Test Plan:
**New Tests added**
- `WideColumnsHelperTest::DumpWideColumns`
- `WideColumnsHelperTest::DumpSliceAsWideColumns`
**Changes added to existing tests**
- `ExternalSSTFileTest::BasicMixed` added to cover mixed case (This test should have been added in https://github.com/facebook/rocksdb/issues/11688). This test does not verify the ldb or sst_dump output. This test was used to create test SST files having some rows with wide columns and some without and the generated SST files were used to manually test sst_dump_tool.
- `createSST()` in `sst_dump_test` now takes `wide_column_one_in` to add wide column value in SST
**dump_wal**
```
./ldb dump_wal --walfile=/tmp/rocksdbtest-226125/db_wide_basic_test_2675429_2308393776696827948/000004.log --print_value --header
```
```
Sequence,Count,ByteSize,Physical Offset,Key(s) : value
1,1,59,0,PUT_ENTITY(0) : 0x:0x68656C6C6F 0x617474725F6E616D6531:0x666F6F 0x617474725F6E616D6532:0x626172
2,1,34,42,PUT_ENTITY(0) : 0x617474725F6F6E65:0x74776F 0x617474725F7468726565:0x666F7572
3,1,17,7d,PUT(0) : 0x7468697264 : 0x62617A
```
**idump**
```
./ldb --db=/tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/ idump
```
```
'first' seq:1, type:22 => :hello attr_name1:foo attr_name2:bar
'second' seq:2, type:22 => attr_one:two attr_three:four
'third' seq:3, type:1 => baz
Internal keys in range: 3
```
**SST Dump from dump_live_files**
```
./ldb --db=/tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/ compact
./ldb --db=/tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/ dump_live_files
```
```
...
==============================
SST Files
==============================
/tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/000013.sst level:1
------------------------------
Process /tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/000013.sst
Sst file format: block-based
'first' seq:0, type:22 => :hello attr_name1:foo attr_name2:bar
'second' seq:0, type:22 => attr_one:two attr_three:four
'third' seq:0, type:1 => baz
...
```
**dump**
```
./ldb --db=/tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/ dump
```
```
first ==> :hello attr_name1:foo attr_name2:bar
second ==> attr_one:two attr_three:four
third ==> baz
Keys in range: 3
```
**scan**
```
./ldb --db=/tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/ scan
```
```
first : :hello attr_name1:foo attr_name2:bar
second : attr_one:two attr_three:four
third : baz
```
**sst_dump**
```
./sst_dump --file=/tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/000013.sst --command=scan
```
```
options.env is 0x7ff54b296000
Process /tmp/rocksdbtest-226125/db_wide_basic_test_3481961_2308393776696827948/000013.sst
Sst file format: block-based
from [] to []
'first' seq:0, type:22 => :hello attr_name1:foo attr_name2:bar
'second' seq:0, type:22 => attr_one:two attr_three:four
'third' seq:0, type:1 => baz
```
Reviewed By: ltamasi
Differential Revision: D48837999
Pulled By: jaykorean
fbshipit-source-id: b0280f0589d2b9716bb9b50530ffcabb397d140f
Summary:
This PR adds a missing piece for the UDT in memtable only feature, which is to automatically increase `full_history_ts_low` when flush happens during recovery.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11774
Test Plan:
Added unit test
make all check
Reviewed By: ltamasi
Differential Revision: D48799109
Pulled By: jowlyzhang
fbshipit-source-id: fd681ed66d9d40904ca2c919b2618eb692686035
Summary:
the value of `done` is always false here, so the sub-condition `!done` will always be true and the check can be removed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11746
Reviewed By: anand1976
Differential Revision: D48656845
Pulled By: ajkr
fbshipit-source-id: 523ba3d07b3af7880c8c8ccb20442fd7c0f49417
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
Summary:
when a key is recorded for locking in a pessimistic transaction, the key is first looked up in a map, and then inserted into the map if it was not already contained.
this can be simplified to an unconditional insert. in the ideal case that all keys are unique, this saves all the find() operations.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11743
Reviewed By: anand1976
Differential Revision: D48656798
Pulled By: ajkr
fbshipit-source-id: d0150de2db757e0c05e1797cfc24380790c71276
Summary:
The user-defined timestamps feature only enforces that for the same key, user-defined timestamps should be non-decreasing. For the user-defined timestamps in memtable only feature, during flush, we check the user-defined timestamps in each memtable to examine if the data is considered expired with regard to `full_history_ts_low`. In this process, it's assuming that a newer memtable should not have smaller user-defined timestamps than an older memtable. This check however is enforcing ordering of user-defined timestamps across keys, as apposed to the vanilla UDT feature, that only enforce ordering of user-defined timestamps for the same key.
This more strict user-defined timestamp ordering requirement could be an issue for secondary instances where commits can be out of order. And after thinking more about it, this requirement is really an overkill to keep the invariants of `full_history_ts_low` which are:
1) users cannot read below `full_history_ts_low`
2) users cannot write at or below `full_history_ts_low`
3) `full_history_ts_low` can only be increasing
As long as RocksDB enforces these 3 checks, we can prohibit inconsistent read that returns a different value. And these three checks are covered in existing APIs.
So this PR removes the extra checks in the UDT in memtable only feature that requires user-defined timestamps to be non decreasing across keys.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11732
Reviewed By: ltamasi
Differential Revision: D48541466
Pulled By: jowlyzhang
fbshipit-source-id: 95453c6e391cbd511c0feab05f0b11c312d17186
Summary:
Fix seg fault in auto_readahead_size with async_io when readahead_size = 0. If readahead_size is trimmed and is 0, it's not eligible for further prefetching and should return.
Error occured when the first buffer already contains data and it goes for prefetching in second buffer leading to assertion failure -
`assert(roundup_len1 >= alignment);
`
because roundup_len1 = length + readahead_size.
length is 0 and readahead_size is also 0.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11769
Test Plan: Reproducible with db_stress with async_io enabled.
Reviewed By: anand1976
Differential Revision: D48743031
Pulled By: akankshamahajan15
fbshipit-source-id: 0e08c41f862f6287ca223fbfaf6cd42fc97b3c87
Summary:
`VersionBuilderMap` type alias definition seem unused.
If this PR can be compiled fine then the alias is probably not needed anymore.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11286
Reviewed By: jaykorean
Differential Revision: D48656747
Pulled By: ajkr
fbshipit-source-id: ac8554922aead7dc3d24fe7e6544a4622578c514
Summary: Python3 makes the use of `(object)` in class inheritance unnecessary. Let's modernize our code by eliminating this.
Reviewed By: itamaro
Differential Revision: D48673915
fbshipit-source-id: a1a6ae8572271eb2898b748c8216ea68e362f06a
Summary:
Fix seg fault in auto_readahead_size
```
db_stress:
internal_repo_rocksdb/repo/table/block_based/partitioned_index_iterator.h:70: virtual rocksdb::IndexValue rocksdb::PartitionedIndexIterator::value() const: Assertion `Valid()' failed.
```
During seek, after calculating readahead_size, db_stress can inject IOError resulting in failure to index_iter_->Seek and making index_iter_ invalid.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11761
Test Plan: Reproducible locally and passed with this fix
Reviewed By: anand1976
Differential Revision: D48696248
Pulled By: akankshamahajan15
fbshipit-source-id: 2be43bf56ad0fc2f95f9093c19c9a1b15a716091
Summary:
* Add some options to cache_bench to use JemallocNodumpAllocator
* Make num_shard_bits option use and report cache-specific defaults
* Add a usleep option to sleep between operations, for simulating a workload with more CPU idle/wait time.
* Use const& for JemallocAllocatorOptions, to improve API usability (e.g. can bind to temporary `{}`)
* InstallStackTraceHandler()
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11758
Test Plan: manual
Reviewed By: jowlyzhang
Differential Revision: D48668479
Pulled By: pdillinger
fbshipit-source-id: b6032fbe09444cdb8f1443a5e017d2eea4f6205a
Summary:
Same as title
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11729
Test Plan: make crash_test -j32
Reviewed By: anand1976
Differential Revision: D48534820
Pulled By: akankshamahajan15
fbshipit-source-id: 3a2a28af98dfad164b82ddaaf9fddb94c53a652e
Summary:
In blackbox tests, db_stress command always run with timeout. Timeout can happen during validation, leaving some of the keys not checked. Since key validation is done in order, it is quite likely that keys those are towards to the end of the set are never validated. This PR adds a final execution, without timeout, to ensure validation is executed for all keys, at least once.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11592
Reviewed By: cbi42
Differential Revision: D48003998
Pulled By: hx235
fbshipit-source-id: 72543475a932f12cf0f57534b7e3b6e07e87080f
Summary:
**Context/Summary:**
Same intention as https://github.com/facebook/rocksdb/pull/2693 - basically we now pick from the last sorted run and expand forward till we can't
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11740
Test Plan:
Existing UT
Stress test
Reviewed By: ajkr
Differential Revision: D48586475
Pulled By: hx235
fbshipit-source-id: 3eb3c3ee1d5f7e0b0d6d649baaeb8c6990fee398
Summary:
Add a bunch of C API functions to expose new `WaitForCompact` function and related options.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11737
Test Plan: unit tests
Reviewed By: jaykorean
Differential Revision: D48568239
Pulled By: abulimov
fbshipit-source-id: 1ff35972d7abacd7e1e17fe2ada1e20cdc88d8de
Summary:
Currently the stress test does not support restoring expected state (to a specific sequence number) when there is unsynced data loss during the reopen phase. This causes a few internal stress test failure with errors like inconsistent value. This PR disables dropping unsynced data during reopen to avoid failures due to this issue. We can re-enable later after we decide to support unsynced data loss during DB reopen in stress test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11731
Test Plan:
* Running this test a few times can fail for inconsistent value before this change
```
./db_stress --acquire_snapshot_one_in=10000 --adaptive_readahead=1 --allow_concurrent_memtable_write=1 --allow_data_in_errors=True --async_io=0 --avoid_flush_during_recovery=0 --avoid_unnecessary_blocking_io=0 --backup_max_size=104857600 --backup_one_in=0 --batch_protection_bytes_per_key=0 --block_protection_bytes_per_key=8 --block_size=16384 --bloom_bits=20.57166126835524 --bottommost_compression_type=disable --bytes_per_sync=262144 --cache_index_and_filter_blocks=1 --cache_size=8388608 --cache_type=auto_hyper_clock_cache --charge_compression_dictionary_building_buffer=1 --charge_file_metadata=1 --charge_filter_construction=0 --charge_table_reader=1 --checkpoint_one_in=0 --checksum_type=kxxHash --clear_column_family_one_in=0 --column_families=1 --compact_files_one_in=1000000 --compact_range_one_in=1000000 --compaction_pri=3 --compaction_style=1 --compaction_ttl=100 --compression_max_dict_buffer_bytes=0 --compression_max_dict_bytes=0 --compression_parallel_threads=1 --compression_type=zstd --compression_use_zstd_dict_trainer=1 --compression_zstd_max_train_bytes=0 --continuous_verification_interval=0 --data_block_index_type=0 --db=/dev/shm/rocksdb_test/rocksdb_crashtest_whitebox --db_write_buffer_size=0 --delpercent=4 --delrangepercent=1 --destroy_db_initially=0 --detect_filter_construct_corruption=0 --disable_wal=0 --enable_compaction_filter=0 --enable_pipelined_write=1 --enable_thread_tracking=0 --expected_values_dir=/dev/shm/rocksdb_test/rocksdb_crashtest_expected --fail_if_options_file_error=1 --fifo_allow_compaction=1 --file_checksum_impl=big --flush_one_in=1000000 --format_version=3 --get_current_wal_file_one_in=0 --get_live_files_one_in=1000000 --get_property_one_in=1000000 --get_sorted_wal_files_one_in=0 --index_block_restart_interval=6 --index_type=3 --ingest_external_file_one_in=0 --initial_auto_readahead_size=16384 --iterpercent=10 --key_len_percent_dist=1,30,69 --level_compaction_dynamic_level_bytes=1 --lock_wal_one_in=1000000 --log2_keys_per_lock=10 --long_running_snapshots=1 --manual_wal_flush_one_in=1000000 --mark_for_compaction_one_file_in=10 --max_auto_readahead_size=0 --max_background_compactions=1 --max_bytes_for_level_base=67108864 --max_key=25000000 --max_key_len=3 --max_manifest_file_size=1073741824 --max_write_batch_group_size_bytes=16777216 --max_write_buffer_number=3 --max_write_buffer_size_to_maintain=0 --memtable_max_range_deletions=100 --memtable_prefix_bloom_size_ratio=0 --memtable_protection_bytes_per_key=1 --memtable_whole_key_filtering=0 --memtablerep=skip_list --min_write_buffer_number_to_merge=2 --mmap_read=0 --mock_direct_io=False --nooverwritepercent=1 --num_file_reads_for_auto_readahead=1 --open_files=-1 --open_metadata_write_fault_one_in=0 --open_read_fault_one_in=5 --open_write_fault_one_in=0 --ops_per_thread=200000 --optimize_filters_for_memory=0 --paranoid_file_checks=1 --partition_filters=0 --partition_pinning=3 --pause_background_one_in=1000000 --periodic_compaction_seconds=10 --prefix_size=-1 --prefixpercent=0 --prepopulate_block_cache=1 --preserve_internal_time_seconds=0 --progress_reports=0 --read_fault_one_in=1000 --readahead_size=524288 --readpercent=50 --recycle_log_file_num=0 --reopen=20 --ribbon_starting_level=0 --secondary_cache_fault_one_in=32 --snapshot_hold_ops=100000 --sst_file_manager_bytes_per_sec=0 --sst_file_manager_bytes_per_truncate=0 --stats_dump_period_sec=10 --subcompactions=3 --sync=0 --sync_fault_injection=1 --target_file_size_base=16777216 --target_file_size_multiplier=1 --test_batches_snapshots=0 --top_level_index_pinning=2 --unpartitioned_pinning=1 --use_direct_io_for_flush_and_compaction=0 --use_direct_reads=0 --use_full_merge_v1=0 --use_get_entity=1 --use_merge=0 --use_multi_get_entity=0 --use_multiget=1 --use_put_entity_one_in=1 --user_timestamp_size=0 --value_size_mult=32 --verify_checksum=1 --verify_checksum_one_in=1000000 --verify_db_one_in=100000 --verify_file_checksums_one_in=1000000 --verify_iterator_with_expected_state_one_in=5 --verify_sst_unique_id_in_manifest=1 --wal_bytes_per_sync=524288 --wal_compression=zstd --write_buffer_size=33554432 --write_dbid_to_manifest=1 --writepercent=35```
Reviewed By: hx235
Differential Revision: D48537494
Pulled By: cbi42
fbshipit-source-id: ddae21b9bb6ee8d67229121f58513e95f7ef6d8d
Summary:
For some ldb commands that doesn't need to open the DB, it's still useful to use the DB's existing OPTIONS file if it's available.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11721
Reviewed By: pdillinger
Differential Revision: D48485540
Pulled By: jowlyzhang
fbshipit-source-id: 2d2db837523044066f1a2c4b59a5c03f6cd35e6b
Summary:
This piggy back the existing last level file temperature statistics test to test the default temperature becoming effective.
While adding this unit test, I found that the approach to swap out and use default temperature in `VersionBuilder::LoadTableHandlers` will miss the L0 files created from flush, and only work for existing SST files, SST files created by compaction. So this PR moves that logic to `TableCache::GetTableReader`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11722
Test Plan:
```
./db_test2 --gtest_filter="*LastLevelStatistics*"
make all check
```
Reviewed By: pdillinger
Differential Revision: D48489171
Pulled By: jowlyzhang
fbshipit-source-id: ac29f7d484916f3218729594c5bb35c4f2979ac2
Summary:
**Context/Summary:** as title, should be harmless. And it's a guessed fix to https://github.com/facebook/rocksdb/issues/11717 while no repro has obtained on my end yet.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11720
Test Plan: existing tests
Reviewed By: cbi42
Differential Revision: D48475661
Pulled By: hx235
fbshipit-source-id: 7c7390319f094c540e703fe2e78a8d601b7a894b
Summary:
Implement trimming of readahead_size under a new option ReadOptions.auto_readahead_size. It'll trim the readahead_size during prefetching upto iterate_upper_bound offset only when ReadOptions.iterate_upper_bound is set, therefore reducing the prefetching of data beyond upper_bound.
It's enabled for both implicit auto readahead size and when ReadOptions.readahead_size is specified and for sync and async_io.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11684
Test Plan: Added new unit test
Reviewed By: anand1976
Differential Revision: D48479723
Pulled By: akankshamahajan15
fbshipit-source-id: 2b1703579caf779105e836b580866ffd7db076fc
Summary:
While it's rare, we may run into a scenario where `WaitForCompact()` waits for background jobs indefinitely. For example, not enough space error will add the job back to the queue while WaitForCompact() waits for _all jobs_ including the jobs that are in the queue to be completed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11711
Test Plan:
`DBCompactionWaitForCompactTest::WaitForCompactToTimeout` added
`timeout` option added to the variables for all of the existing DBCompactionWaitForCompactTests
Reviewed By: pdillinger, jowlyzhang
Differential Revision: D48416390
Pulled By: jaykorean
fbshipit-source-id: 7b6a12f705ab6c6dfaf8ad736a484ca654a86106
Summary:
This PR implements a new admission policy for the compressed secondary cache, which includes the functionality of the existing policy, and also admits items evicted from the primary block cache with the hit bit set. Effectively, the new policy works as follows -
1. When an item is demoted from the primary cache without a hit, a placeholder is inserted in the compressed cache. A second demotion will insert the full entry.
2. When an item is promoted from the compressed cache to the primary cache for the first time, a placeholder is inserted in the primary. The second promotion inserts the full entry, while erasing it form the compressed cache.
3. If an item is demoted from the primary cache with the hit bit set, it is immediately inserted in the compressed secondary cache.
The ```TieredVolatileCacheOptions``` has been updated with a new option, ```adm_policy```, which allows the policy to be selected.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11713
Reviewed By: pdillinger
Differential Revision: D48444512
Pulled By: anand1976
fbshipit-source-id: b4cbf8c169a88097dff08e36e8bc4b3088de1492
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11714
Fixes T161017540.
The staging build starts failing with an undefined symbol error:
```
ld.lld: error: undefined symbol: std::enable_if<rocksdb::ParsedFullFilterBlock::kCacheEntryRole == (rocksdb::CacheEntryRole)13 || true, rocksdb::Status>::type rocksdb::BlockBasedTable::MaybeReadBlockAndLoadToCache<rocksdb::ParsedFullFilterBlock>(rocksdb::FilePrefetchBuffer*, rocksdb::ReadOptions const&, rocksdb::BlockHandle const&, rocksdb::UncompressionDict const&, bool, rocksdb::CachableEntry<rocksdb::ParsedFullFilterBlock>*, rocksdb::GetContext*, rocksdb::BlockCacheLookupContext*, rocksdb::BlockContents*, bool) const
```
This is the `MaybeReadBlockAndLoadToCache` function where `TBlocklike = ParsedFullFilterBlock`. The trigger was an FDO profile update D48261413.
`MaybeReadBlockAndLoadToCache` is used in the same translation unit `block_based_table_reader.cc`, and also in another file `partitioned_filter_block.cc`. The later was the file that couldn't find the symbol. It seems after the FDO profile update, `MaybeReadBlockAndLoadToCache` may've got inlined into its caller in `block_based_table_reader.cc`. And with no knowledge of other usages, the symbol got stripped.
Explicitly instantiate the template similar to how `RetrieveBlock` was handled.
Reviewed By: pdillinger, akankshamahajan15
Differential Revision: D48400574
fbshipit-source-id: d4a80999bfb6ce4afa80678444139fcd8ae84aa4
Summary:
Add a column family option `default_temperature` that will be used for file reading accounting purpose, such as io statistics, for files that don't have an explicitly set temperature.
This options is not a mutable one, changing its value would require a DB restart. This is to avoid the confusion that had the option being a mutable one, the users may expect it to take effect on all files immediately, while in reality, it would only become effective for SST files opened in the future.
This `default_temperature` also just affect accounting during one DB session. It won't be recorded in manifest as the file's temperature and can be different across different DB sessions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11708
Test Plan:
```
make all check
```
Reviewed By: pdillinger
Differential Revision: D48375763
Pulled By: jowlyzhang
fbshipit-source-id: eb756696c14a694c6e2a93d2bb6f040563194981
Summary:
* JemallocNodumpAllocator was passing a size_t to FastRange32, which could cause compilation errors or warnings (seen with clang)
* Fixed the order of arguments to match what would be used with modulo operator (%), for clarity.
Fixes https://github.com/facebook/rocksdb/issues/11006
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11707
Test Plan: no functional change, existing tests
Reviewed By: ajkr
Differential Revision: D48435149
Pulled By: pdillinger
fbshipit-source-id: e6e8b107ded4eceda37db20df59985c846a2546b
Summary:
For leveled compaction, RocksDB has a special kind of compaction with reason "kBottommmostFiles" that compacts bottommost level files to clear data held by snapshots (more detail in https://github.com/facebook/rocksdb/issues/3009). Such compactions can happen soon after a relevant snapshot is released. For some use cases, a bottommost file may contain only a small amount of keys that can be cleared, so compacting such a file has a high write amp. In addition, these bottommost files may be compacted in compactions with reason other than "kBottommmostFiles" if we wait for some time (so that enough data is ingested to trigger such a compaction). This PR introduces an option `bottommost_file_compaction_delay` to specify the delay of these bottommost level single file compactions.
* The main change is in `VersionStorageInfo::ComputeBottommostFilesMarkedForCompaction()` where we only add a file to `bottommost_files_marked_for_compaction_` if it oldest_snapshot is larger than its non-zero largest_seqno **and** the file is old enough. Note that if a file is not old enough but its largest_seqno is less than oldest_snapshot, we exclude it from the calculation of `bottommost_files_mark_threshold_`. This makes the change simpler, but such a file's eligibility for compaction will only be checked the next time `ComputeBottommostFilesMarkedForCompaction()` is called. This happens when a new Version is created (compaction, flush, SetOptions()...), a new enough snapshot is released (`VersionStorageInfo::UpdateOldestSnapshot()`) or when a compaction is picked and compaction score has to be re-calculated.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11701
Test Plan:
* Add two unit tests to test when bottommost_file_compaction_delay > 0.
* Ran crash test with the new option.
Reviewed By: jaykorean, ajkr
Differential Revision: D48331564
Pulled By: cbi42
fbshipit-source-id: c584f3dc5f6354fce3ed65f4c6366dc450b15ba8
Summary:
As titled. User-defined timestamp feature users sometimes directly call the user comparator to do validation on their side too. Having access to the root comparator can help make their code consistent for when UDT is enabled and disabled.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11704
Reviewed By: ltamasi
Differential Revision: D48355090
Pulled By: jowlyzhang
fbshipit-source-id: 26bc73543bfb379ef548d1361803d6f8c308cef6
Summary:
As titled, mostly adding documentation. While updating one usage of these util functions in the external file ingestion job based on code inspection.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11674
Test Plan:
```
make check
```
Note that no unit test was added or updated to check the change in the external file ingestion flow works. This is because user-defined timestamp doesn't support bulk loading yet. There could be other missing pieces that are needed to make this flow functional and testable. That work is separately tracked and unit tests will be added then.
Reviewed By: cbi42
Differential Revision: D48271338
Pulled By: jowlyzhang
fbshipit-source-id: c05c3440f1c08632dd0de51b563a30b44b4eb8b5
Summary:
It's a bit repetitive in order to give reasonably informative error messages.
I also removed total_order_seek in cases where it's not needed, just to make sure a case that shouldn't matter really doesn't.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11698
Test Plan:
run it -
```
$ DEBUG_LEVEL=0 TEST_TMPDIR=/dev/shm python3 tools/db_crashtest.py blackbox --max_key=100000 --duration=86400 --interval=10 --write_buffer_size=524288 --target_file_size_base=524288 --max_bytes_for_level_base=2097152 --compression_type=none --blob_compression_type=none --writepercent=50 -iterpercent=45 -readpercent=0 -prefixpercent=0 --prefix_size=0 --verify_iterator_with_expected_state_one_in=10 --test_batches_snapshots=0 -enable_compaction_filter=0
```
Reviewed By: cbi42
Differential Revision: D48285036
Pulled By: ajkr
fbshipit-source-id: 51b147bd7c8011740629ae2fd8114d3d48ce7137
Summary:
* The plan is for AutoHyperClockCache to be selected when HyperClockCacheOptions::estimated_entry_charge == 0, and in that case to use a new configuration option min_avg_entry_charge for determining an extreme case maximum size for the hash table. For the placeholder, a hack is in place in HyperClockCacheOptions::MakeSharedCache() to make the unit tests happy despite the new options not really making sense with the current implementation.
* Mostly updating and refactoring tests to test both the current HCC (internal name FixedHyperClockCache) and a placeholder for the new version (internal name AutoHyperClockCache).
* Simplify some existing tests not to depend directly on cache type.
* Type-parameterize the shard-level unit tests, which unfortunately requires more syntax like `this->` in places for disambiguation.
* Added means of choosing auto_hyper_clock_cache to cache_bench, db_bench, and db_stress, including add to crash test.
* Add another templated class BaseHyperClockCache to reduce future copy-paste
* Added ReportProblems support to cache_bench
* Added a DEBUG-level diagnostic to ReportProblems for the variance in load factor throughout the table, which will become more of a concern with linear hashing to be used in the Auto implementation. Example with current Fixed HCC:
```
2023/08/10-13:41:41.602450 6ac36 [DEBUG] [che/clock_cache.cc:1507] Slot occupancy stats: Overall 49% (129008/262144), Min/Max/Window = 39%/60%/500, MaxRun{Pos/Neg} = 18/17
```
In other words, with overall occupancy of 49%, the lowest across any 500 contiguous cells is 39% and highest 60%. Longest run of occupied is 18 and longest run of unoccupied is 17. This seems consistent with random samples from a uniform distribution.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11692
Test Plan: Shouldn't be any meaningful changes yet to production code or to what is tested, but there is temporary redundancy in testing until the new implementation is plugged in.
Reviewed By: jowlyzhang
Differential Revision: D48247413
Pulled By: pdillinger
fbshipit-source-id: 11541f996d97af403c2e43c92fb67ff22dd0b5da
Peter Dillinger