Summary:
Same as title
Error:
```
block_cache_trace_analyzer: ./db/dbformat.h:421: uint64_t rocksdb::GetInternalKeySeqno(const rocksdb::Slice&): Assertion `n >= kNumInternalBytes' failed.
Aborted (core dumped)
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11042
Test Plan:
- Added new unit test which fails without the fix.
- Also ran manually on traces to confirm.
Reviewed By: anand1976
Differential Revision: D42481587
Pulled By: akankshamahajan15
fbshipit-source-id: 7c33eb03a4a4d8ffbabcfbe0efa1e4d11bde3ba2
Summary:
When running `-benchmarks=compact`, `-subcompactions` does not take effect.
`-subcompactions` option comment says it is for L0-L1 compactions, it is natural to extend it to CompactionRangeOptions.max_subcompactions.
This PR set CompactionRangeOptions.max_subcompactions = FLAGS_subcompactions
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11077
Reviewed By: akankshamahajan15
Differential Revision: D42506251
Pulled By: ajkr
fbshipit-source-id: f77c9a99d32ff7af59f3c452c9e16aaeb0360304
Summary:
**Context/Summary:**
This reverts commit fc74abb436 and related HISTORY record.
The issue with PR 10777 or general approach using earliest_mem_seqno like https://github.com/facebook/rocksdb/pull/5958#issue-511150930 is that the earliest seqno of memtable of each CFs does not get persisted and will always start with 0 upon Recover(). Later when creating a new memtable in certain CF, we use the last seqno of the whole DB (but not of that CF from previous DB session) for this CF. This will lead to false positive overlapping seqno and PR 10777 will throw something like https://github.com/facebook/rocksdb/blob/main/db/compaction/compaction_picker.cc#L1002-L1004
Luckily a more elegant and complete solution to the overlapping seqno problem these PR aim to solve does not have above problem, see https://github.com/facebook/rocksdb/pull/10922. It is already being pursued and in the process of review. So we can just revert this PR and focus on getting PR10922 to land.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10999
Test Plan: make check
Reviewed By: anand1976
Differential Revision: D41572604
Pulled By: hx235
fbshipit-source-id: 9d9bdf594abd235e2137045cef513ca0b14e0a3a
Summary:
This was just a stepping stone to what eventually became HyperClockCache, and is now just more code to maintain.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10954
Test Plan: tests updated
Reviewed By: akankshamahajan15
Differential Revision: D41310123
Pulled By: pdillinger
fbshipit-source-id: 618ee148a1a0a29ee756ba8fe28359617b7cd67c
Summary:
This option allows a custom cleanup command line for a non-Posix file system to be used by db_crashtest.py to cleanup between runs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10883
Test Plan: Run the whitebox crash test
Reviewed By: pdillinger
Differential Revision: D40726424
Pulled By: anand1976
fbshipit-source-id: b827f6b583ff78f9ca75ced2d96f7e58f5200432
Summary:
Some lines of .h and .cc files are not properly fomatted. Clear them up with clang format.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10868
Test Plan: Watch existing CI to pass
Reviewed By: ajkr
Differential Revision: D40683485
fbshipit-source-id: 491fbb78b2cdcb948164f306829909ad816d5d0b
Summary:
**Context:**
Same as https://github.com/facebook/rocksdb/pull/5958#issue-511150930 but apply the fix to FIFO Compaction case
Repro:
```
COERCE_CONTEXT_SWICH=1 make -j56 db_stress
./db_stress --acquire_snapshot_one_in=0 --adaptive_readahead=0 --allow_data_in_errors=True --async_io=1 --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_size=16384 --bloom_bits=18 --bottommost_compression_type=disable --bytes_per_sync=262144 --cache_index_and_filter_blocks=0 --cache_size=8388608 --cache_type=lru_cache --charge_compression_dictionary_building_buffer=0 --charge_file_metadata=1 --charge_filter_construction=1 --charge_table_reader=1 --checkpoint_one_in=0 --checksum_type=kCRC32c --clear_column_family_one_in=0 --column_families=1 --compact_files_one_in=0 --compact_range_one_in=1000 --compaction_pri=3 --open_files=-1 --compaction_style=2 --fifo_allow_compaction=1 --compaction_ttl=0 --compression_max_dict_buffer_bytes=8388607 --compression_max_dict_bytes=16384 --compression_parallel_threads=1 --compression_type=zlib --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_test0/rocksdb_crashtest_whitebox --db_write_buffer_size=8388608 --delpercent=4 --delrangepercent=1 --destroy_db_initially=1 --detect_filter_construct_corruption=0 --disable_wal=0 --enable_compaction_filter=0 --enable_pipelined_write=1 --fail_if_options_file_error=1 --file_checksum_impl=none --flush_one_in=1000 --format_version=5 --get_current_wal_file_one_in=0 --get_live_files_one_in=0 --get_property_one_in=0 --get_sorted_wal_files_one_in=0 --index_block_restart_interval=15 --index_type=3 --ingest_external_file_one_in=100 --initial_auto_readahead_size=0 --iterpercent=10 --key_len_percent_dist=1,30,69 --level_compaction_dynamic_level_bytes=True --log2_keys_per_lock=10 --long_running_snapshots=0 --mark_for_compaction_one_file_in=10 --max_auto_readahead_size=16384 --max_background_compactions=20 --max_bytes_for_level_base=10485760 --max_key=100000 --max_key_len=3 --max_manifest_file_size=1073741824 --max_write_batch_group_size_bytes=1048576 --max_write_buffer_number=3 --max_write_buffer_size_to_maintain=4194304 --memtable_prefix_bloom_size_ratio=0.5 --memtable_protection_bytes_per_key=1 --memtable_whole_key_filtering=1 --memtablerep=skip_list --mmap_read=1 --mock_direct_io=False --nooverwritepercent=1 --num_file_reads_for_auto_readahead=0 --num_levels=1 --open_metadata_write_fault_one_in=0 --open_read_fault_one_in=32 --open_write_fault_one_in=0 --ops_per_thread=200000 --optimize_filters_for_memory=0 --paranoid_file_checks=1 --partition_filters=0 --partition_pinning=1 --pause_background_one_in=0 --periodic_compaction_seconds=0 --prefix_size=8 --prefixpercent=5 --prepopulate_block_cache=0 --progress_reports=0 --read_fault_one_in=0 --readahead_size=16384 --readpercent=45 --recycle_log_file_num=1 --reopen=20 --ribbon_starting_level=999 --snapshot_hold_ops=1000 --sst_file_manager_bytes_per_sec=0 --sst_file_manager_bytes_per_truncate=0 --subcompactions=2 --sync=0 --sync_fault_injection=0 --target_file_size_base=524288 --target_file_size_multiplier=2 --test_batches_snapshots=0 --top_level_index_pinning=3 --unpartitioned_pinning=0 --use_direct_io_for_flush_and_compaction=0 --use_direct_reads=0 --use_full_merge_v1=1 --use_merge=0 --use_multiget=1 --user_timestamp_size=0 --value_size_mult=32 --verify_checksum=1 --verify_checksum_one_in=0 --verify_db_one_in=1000 --verify_sst_unique_id_in_manifest=1 --wal_bytes_per_sync=0 --wal_compression=zstd --write_buffer_size=524288 --write_dbid_to_manifest=0 --writepercent=35
put or merge error: Corruption: force_consistency_checks(DEBUG): VersionBuilder: L0 file https://github.com/facebook/rocksdb/issues/479 with seqno 23711 29070 vs. file https://github.com/facebook/rocksdb/issues/482 with seqno 27138 29049
```
**Summary:**
FIFO only does intra-L0 compaction in the following four cases. For other cases, FIFO drops data instead of compacting on data, which is irrelevant to the overlapping seqno issue we are solving.
- [FIFOCompactionPicker::PickSizeCompaction](https://github.com/facebook/rocksdb/blob/7.6.fb/db/compaction/compaction_picker_fifo.cc#L155) when `total size < compaction_options_fifo.max_table_files_size` and `compaction_options_fifo.allow_compaction == true`
- For this path, we simply reuse the fix in `FindIntraL0Compaction` https://github.com/facebook/rocksdb/pull/5958/files#diff-c261f77d6dd2134333c4a955c311cf4a196a08d3c2bb6ce24fd6801407877c89R56
- This path was not stress-tested at all. Therefore we covered `fifo.allow_compaction` in stress test to surface the overlapping seqno issue we are fixing here.
- [FIFOCompactionPicker::PickCompactionToWarm](https://github.com/facebook/rocksdb/blob/7.6.fb/db/compaction/compaction_picker_fifo.cc#L313) when `compaction_options_fifo.age_for_warm > 0`
- For this path, we simply replicate the idea in https://github.com/facebook/rocksdb/pull/5958#issue-511150930 and skip files of largest seqno greater than `earliest_mem_seqno`
- This path was not stress-tested at all. However covering `age_for_warm` option worths a separate PR to deal with db stress compatibility. Therefore we manually tested this path for this PR
- [FIFOCompactionPicker::CompactRange](https://github.com/facebook/rocksdb/blob/7.6.fb/db/compaction/compaction_picker_fifo.cc#L365) that ends up picking one of the above two compactions
- [CompactionPicker::CompactFiles](https://github.com/facebook/rocksdb/blob/7.6.fb/db/compaction/compaction_picker.cc#L378)
- Since `SanitizeCompactionInputFiles()` will be called [before](https://github.com/facebook/rocksdb/blob/7.6.fb/db/compaction/compaction_picker.h#L111-L113) `CompactionPicker::CompactFiles` , we simply replicate the idea in https://github.com/facebook/rocksdb/pull/5958#issue-511150930 in `SanitizeCompactionInputFiles()`. To simplify implementation, we return `Stats::Abort()` on encountering seqno-overlapped file when doing compaction to L0 instead of skipping the file and proceed with the compaction.
Some additional clean-up included in this PR:
- Renamed `earliest_memtable_seqno` to `earliest_mem_seqno` for consistent naming
- Added comment about `earliest_memtable_seqno` in related APIs
- Made parameter `earliest_memtable_seqno` constant and required
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10777
Test Plan:
- make check
- New unit test `TEST_P(DBCompactionTestFIFOCheckConsistencyWithParam, FlushAfterIntraL0CompactionWithIngestedFile)`corresponding to the above 4 cases, which will fail accordingly without the fix
- Regular CI stress run on this PR + stress test with aggressive value https://github.com/facebook/rocksdb/pull/10761 and on FIFO compaction only
Reviewed By: ajkr
Differential Revision: D40090485
Pulled By: hx235
fbshipit-source-id: 52624186952ee7109117788741aeeac86b624a4f
Summary:
Refactor the classes, APIs and data structures for block cache tracing to allow a user provided trace writer to be used. Currently, only a TraceWriter is supported, with a default built-in implementation of FileTraceWriter. The TraceWriter, however, takes a flat trace record and is thus only suitable for file tracing. This PR introduces an abstract BlockCacheTraceWriter class that takes a structured BlockCacheTraceRecord. The BlockCacheTraceWriter implementation can then format and log the record in whatever way it sees fit. The default BlockCacheTraceWriterImpl does file tracing using a user provided TraceWriter.
`DB::StartBlockTrace` will internally redirect to changed `BlockCacheTrace::StartBlockCacheTrace`.
New API `DB::StartBlockTrace` is also added that directly takes `BlockCacheTraceWriter` pointer.
This same philosophy can be applied to KV and IO tracing as well.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10811
Test Plan:
existing unit tests
Old API DB::StartBlockTrace checked with db_bench tool
create database
```
./db_bench --benchmarks="fillseq" \
--key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 \
--cache_index_and_filter_blocks --cache_size=1048576 \
--disable_auto_compactions=1 --disable_wal=1 --compression_type=none \
--min_level_to_compress=-1 --compression_ratio=1 --num=10000000
```
To trace block cache accesses when running readrandom benchmark:
```
./db_bench --benchmarks="readrandom" --use_existing_db --duration=60 \
--key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 \
--cache_index_and_filter_blocks --cache_size=1048576 \
--disable_auto_compactions=1 --disable_wal=1 --compression_type=none \
--min_level_to_compress=-1 --compression_ratio=1 --num=10000000 \
--threads=16 \
-block_cache_trace_file="/tmp/binary_trace_test_example" \
-block_cache_trace_max_trace_file_size_in_bytes=1073741824 \
-block_cache_trace_sampling_frequency=1
```
Reviewed By: anand1976
Differential Revision: D40435289
Pulled By: akankshamahajan15
fbshipit-source-id: fa2755f4788185e19f4605e731641cfd21ab3282
Summary:
Instead of existing calls to ps from gnu_parallel, call a new wrapper that does ps, looks for unit test like processes, and uses pstack or gdb to print thread stack traces. Also, using `ps -wwf` instead of `ps -wf` ensures output is not cut off.
For security, CircleCI runs with security restrictions on ptrace (/proc/sys/kernel/yama/ptrace_scope = 1), and this change adds a work-around to `InstallStackTraceHandler()` (only used by testing tools) to allow any process from the same user to debug it. (I've also touched >100 files to ensure all the unit tests call this function.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10828
Test Plan: local manual + temporary infinite loop in a unit test to observe in CircleCI
Reviewed By: hx235
Differential Revision: D40447634
Pulled By: pdillinger
fbshipit-source-id: 718a4c4a5b54fa0f9af2d01a446162b45e5e84e1
Summary:
This is a small follow-up to https://github.com/facebook/rocksdb/pull/10821. The goal of that PR was to hold `test_batches_snapshots` fixed across all `db_stress` invocations; however, that patch didn't address the case when `test_batches_snapshots` is unset due to a conflicting `enable_compaction_filter` or `prefix_size` setting. This PR updates the logic so the other parameter is sanitized instead in the case of such conflicts.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10830
Reviewed By: riversand963
Differential Revision: D40444548
Pulled By: ltamasi
fbshipit-source-id: 0331265704904b729262adec37139292fcbb7805
Summary:
We have recently made some stress test improvements that rely on decoding the "value base" from the values stored in the database. This logic does not currently support the case when some KVs are written by a non-batched ops run and some by a batched ops run. The patch temporarily disables mixing these two.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10821
Reviewed By: riversand963
Differential Revision: D40367326
Pulled By: ltamasi
fbshipit-source-id: 66f2e0cbc097ab6b1f9e4b39b833bd466f1aaab5
Summary:
This has several small improvements.
benchmark.sh
* add BYTES_PER_SYNC as an env variable
* use --prepopulate_block_cache when O_DIRECT is used
* use --undefok to list options that don't work for all 7.x releases
* print "failure" in report.tsv when a benchmark fails
* parse the slightly different throughput line used by db_bench for multireadrandom
* remove the trailing comma for BlobDB size before printing it in report.tsv
* use the last line of the output from /bin/time as there can be more than one line when db_bench has a non-zero exit
* fix more bash lint warnings
* add ",stats" to the --benchmark=... lines to get stats at the end of each benchmark
benchmark_compare.sh
* run revrange immediately after fillseq to let compaction debt get removed
* add --multiread_batched when --benchmarks=multireadrandom is used
* use --benchmarks=overwriteandwait when supported to get a more accurate measure of write-amp
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10803
Test Plan: Run it for leveled, universal and BlobDB
Reviewed By: jay-zhuang
Differential Revision: D40278315
Pulled By: mdcallag
fbshipit-source-id: 793134ddc7d48d05a07436cd8942c375a23983a7
Summary:
This change is motivated by ensuring that `ldb update_manifest` or `UpdateManifestForFilesState` can run without expecting files to open when the old temperature is provided (in case the FileSystem strictly interprets non-kUnknown), but ended up fixing a problem in `OfflineManifestWriter` (used by `ldb unsafe_remove_sst_file`) where it would open some SST files during recovery and expect them to match the prior manifest state, even if not required by the intended new state.
Also update BackupEngine to retry with Temperature kUnknown when reading file with potentially "wrong" temperature.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10796
Test Plan: tests added/updated, that fail before the change(s) and now pass
Reviewed By: jay-zhuang
Differential Revision: D40232645
Pulled By: jay-zhuang
fbshipit-source-id: b5aa2688aecfe0c320b80a7da689b315414c20be
Summary:
We have seen some rare crash test failures in HyperClockCache, and the source could certainly be a bug fixed in this change, in ClockHandleTable::ConstApplyToEntriesRange. It wasn't properly accounting for the fact that incrementing the acquire counter could be ineffective, due to parallel updates. (When incrementing the acquire counter is ineffective, it is incorrect to then decrement it.)
This change includes some other minor clean-up in HyperClockCache, and adds stats_dump_period_sec with a much lower period to the crash test. This should be the primary caller of ApplyToEntries, in collecting cache entry stats.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10768
Test Plan: haven't been able to reproduce the failure, but should be in a better state (bug fix and improved crash test)
Reviewed By: anand1976
Differential Revision: D40034747
Pulled By: anand1976
fbshipit-source-id: a06fcefe146e17ee35001984445cedcf3b63eb68
Summary:
`SstFileWriter` currently does not support the `PutEntity` API, so in `TestIngestExternalFile` all key-values are written using regular `Put`s. This violates the assumption that whether or not a key corresponds to a plain old key-value or a wide-column entity can be determined by solely looking at the "value base" used when generating the value. The patch fixes this issue by disabling ingestion when `PutEntity` is enabled in the stress tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10769
Test Plan: Ran a simple blackbox stress test.
Reviewed By: akankshamahajan15
Differential Revision: D40042132
Pulled By: ltamasi
fbshipit-source-id: 93e75ff55545b7b69fa4ddef1d96093c961158a0
Summary:
Add user-defined timestamp support for range deletion. The new API is `DeleteRange(opt, cf, begin_key, end_key, ts)`. Most of the change is to update the comparator to compare without timestamp. Other than that, major changes are
- internal range tombstone data structures (`FragmentedRangeTombstoneList`, `RangeTombstone`, etc.) to store timestamps.
- Garbage collection of range tombstones and range tombstone covered keys during compaction.
- Get()/MultiGet() to return the timestamp of a range tombstone when needed.
- Get/Iterator with range tombstones bounded by readoptions.timestamp.
- timestamp crash test now issues DeleteRange by default.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10661
Test Plan:
- Added unit test: `make check`
- Stress test: `python3 tools/db_crashtest.py --enable_ts whitebox --readpercent=57 --prefixpercent=4 --writepercent=25 -delpercent=5 --iterpercent=5 --delrangepercent=4`
- Ran `db_bench` to measure regression when timestamp is not enabled. The tests are for write (with some range deletion) and iterate with DB fitting in memory: `./db_bench--benchmarks=fillrandom,seekrandom --writes_per_range_tombstone=200 --max_write_buffer_number=100 --min_write_buffer_number_to_merge=100 --writes=500000 --reads=500000 --seek_nexts=10 --disable_auto_compactions -disable_wal=true --max_num_range_tombstones=1000`. Did not see consistent regression in no timestamp case.
| micros/op | fillrandom | seekrandom |
| --- | --- | --- |
|main| 2.58 |10.96|
|PR 10661| 2.68 |10.63|
Reviewed By: riversand963
Differential Revision: D39441192
Pulled By: cbi42
fbshipit-source-id: f05aca3c41605caf110daf0ff405919f300ddec2
Summary:
**Context/Summary:**
Introduce `manual_wal_flush_one_in` as titled.
- When `manual_wal_flush_one_in > 0`, we also need tracing to correctly verify recovery because WAL data can be lost in this case when `FlushWAL()` is not explicitly called by users of RocksDB (in our case, db stress) and the recovery from such potential WAL data loss is a prefix recovery that requires tracing to verify. As another consequence, we need to disable features can't run under unsync data loss with `manual_wal_flush_one_in`
Incompatibilities fixed along the way:
```
db_stress: db/db_impl/db_impl_open.cc:2063: static rocksdb::Status rocksdb::DBImpl::Open(const rocksdb::DBOptions&, const string&, const std::vector<rocksdb::ColumnFamilyDescriptor>&, std::vector<rocksdb::ColumnFamilyHandle*>*, rocksdb::DB**, bool, bool): Assertion `impl->TEST_WALBufferIsEmpty()' failed.
```
- It turns out that `Writer::AddCompressionTypeRecord` before this assertion `EmitPhysicalRecord(kSetCompressionType, encode.data(), encode.size());` but do not trigger flush if `manual_wal_flush` is set . This leads to `impl->TEST_WALBufferIsEmpty()' is false.
- As suggested, assertion is removed and violation case is handled by `FlushWAL(sync=true)` along with refactoring `TEST_WALBufferIsEmpty()` to be `WALBufferIsEmpty()` since it is used in prod code now.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10698
Test Plan:
- Locally running `python3 tools/db_crashtest.py blackbox --manual_wal_flush_one_in=1 --manual_wal_flush=1 --sync_wal_one_in=100 --atomic_flush=1 --flush_one_in=100 --column_families=3`
- Joined https://github.com/facebook/rocksdb/pull/10624 in auto CI testings with all RocksDB stress/crash test jobs
Reviewed By: ajkr
Differential Revision: D39593752
Pulled By: ajkr
fbshipit-source-id: 3a2135bb792c52d2ffa60257d4fbc557fb04d2ce
Summary:
The patch adds the `PutEntity` API to the non-batched, batched, and
CF consistency stress tests. Namely, when the new `db_stress` command
line parameter `use_put_entity_one_in` is greater than zero, one in
N writes on average is performed using `PutEntity` rather than `Put`.
The wide-column entity written has the generated value in its default
column; in addition, it contains up to three additional columns where
the original generated value is divided up between the column name and the
column value (with the column name containing the first k characters of
the generated value, and the column value containing the rest). Whether
`PutEntity` is used (and if so, how many columns the entity has) is completely
determined by the "value base" used to generate the value (that is, there is
no randomness involved). Assuming the same `use_put_entity_one_in` setting
is used across `db_stress` invocations, this enables us to reconstruct and
validate the entity during subsequent `db_stress` runs.
Note that `PutEntity` is currently incompatible with `Merge`, transactions, and
user-defined timestamps; these combinations are currently disabled/disallowed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760
Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script.
Reviewed By: riversand963
Differential Revision: D39939032
Pulled By: ltamasi
fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
Summary:
An add-on to https://github.com/facebook/rocksdb/pull/6818 to complete adding single-level universal compaction to stress/crash testing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10732
Test Plan:
- Locally run for 10 min `python3 ./tools/db_crashtest.py whitebox --simple --compaction_style=1 --num_levels=1 -max_key=1000000 -value_size_mult=33 -write_buffer_size=524288 -target_file_size_base=524288 -max_bytes_for_level_base=2097152 --duration=120 --interval=10 --ops_per_thread=1000 --random_kill_odd=887`
- Check LOG to confirm single-level universal compaction is called
- Manual testing and log checking to ensure destroy_db_initially=1 is correctly set across runs with different compaction styles (i.e, in the second half of whitebox testing).
- [ongoing]CI jobs stress test
Reviewed By: ajkr
Differential Revision: D39797612
Pulled By: ajkr
fbshipit-source-id: 16f5c40c3464c57360c06c8305f92118e426149c
Summary:
**Context:**
Prior to this PR, correctness testing with un-sync data loss [disabled](https://github.com/facebook/rocksdb/pull/10605) transaction (`use_txn=1`) thus all of the `txn_write_policy` . This PR improved that by adding support for one policy - WriteCommit (`txn_write_policy=0`).
**Summary:**
They key to this support is (a) handle Mark{Begin, End}Prepare/MarkCommit/MarkRollback in constructing ExpectedState under WriteCommit policy correctly and (b) monitor CI jobs and solve any test incompatibility issue till jobs are stable. (b) will be part of the test plan.
For (a)
- During prepare (i.e, between `MarkBeginPrepare()` and `MarkEndPrepare(xid)`), `ExpectedStateTraceRecordHandler` will buffer all writes by adding all writes to an internal `WriteBatch`.
- On `MarkEndPrepare()`, that `WriteBatch` will be associated with the transaction's `xid`.
- During the commit (i.e, on `MarkCommit(xid)`), `ExpectedStateTraceRecordHandler` will retrieve and iterate the internal `WriteBatch` and finally apply those writes to `ExpectedState`
- During the rollback (i.e, on `MarkRollback(xid)`), `ExpectedStateTraceRecordHandler` will erase the internal `WriteBatch` from the map.
For (b) - one major issue described below:
- TransactionsDB in db stress recovers prepared-but-not-committed txns from the previous crashed run by randomly committing or rolling back it at the start of the current run, see a historical [PR](6d06be22c0) predated correctness testing.
- And we will verify those processed keys in a recovered db against their expected state.
- However since now we turn on `sync_fault_injection=1` where the expected state is constructed from the trace instead of using the LATEST.state from previous run. The expected state now used to verify those processed keys won't contain UNKNOWN_SENTINEL as they should - see test 1 for a failed case.
- Therefore, we decided to manually update its expected state to be UNKNOWN_SENTINEL as part of the processing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10624
Test Plan:
1. Test exposed the major issue described above. This test will fail without setting UNKNOWN_SENTINEL in expected state during the processing and pass after
```
db=/dev/shm/rocksdb_crashtest_blackbox
exp=/dev/shm/rocksdb_crashtest_expected
dbt=$db.tmp
expt=$exp.tmp
rm -rf $db $exp
mkdir -p $exp
echo "RUN 1"
./db_stress \
--clear_column_family_one_in=0 --column_families=1 --db=$db --delpercent=10 --delrangepercent=0 --destroy_db_initially=0 --expected_values_dir=$exp --iterpercent=0 --key_len_percent_dist=1,30,69 --max_key=1000000 --max_key_len=3 --prefixpercent=0 --readpercent=0 --reopen=0 --ops_per_thread=100000000 --test_batches_snapshots=0 --value_size_mult=32 --writepercent=90 \
--use_txn=1 --txn_write_policy=0 --sync_fault_injection=1 &
pid=$!
sleep 0.2
sleep 20
kill $pid
sleep 0.2
echo "RUN 2"
./db_stress \
--clear_column_family_one_in=0 --column_families=1 --db=$db --delpercent=10 --delrangepercent=0 --destroy_db_initially=0 --expected_values_dir=$exp --iterpercent=0 --key_len_percent_dist=1,30,69 --max_key=1000000 --max_key_len=3 --prefixpercent=0 --readpercent=0 --reopen=0 --ops_per_thread=100000000 --test_batches_snapshots=0 --value_size_mult=32 --writepercent=90 \
--use_txn=1 --txn_write_policy=0 --sync_fault_injection=1 &
pid=$!
sleep 0.2
sleep 20
kill $pid
sleep 0.2
echo "RUN 3"
./db_stress \
--clear_column_family_one_in=0 --column_families=1 --db=$db --delpercent=10 --delrangepercent=0 --destroy_db_initially=0 --expected_values_dir=$exp --iterpercent=0 --key_len_percent_dist=1,30,69 --max_key=1000000 --max_key_len=3 --prefixpercent=0 --readpercent=0 --reopen=0 --ops_per_thread=100000000 --test_batches_snapshots=0 --value_size_mult=32 --writepercent=90 \
--use_txn=1 --txn_write_policy=0 --sync_fault_injection=1
```
2. Manual testing to ensure ExpectedState is constructed correctly during recovery by verifying it against previously crashed TransactionDB's WAL.
- Run the following command to crash a TransactionDB with WriteCommit policy. Then `./ldb dump_wal` on its WAL file
```
db=/dev/shm/rocksdb_crashtest_blackbox
exp=/dev/shm/rocksdb_crashtest_expected
rm -rf $db $exp
mkdir -p $exp
./db_stress \
--clear_column_family_one_in=0 --column_families=1 --db=$db --delpercent=10 --delrangepercent=0 --destroy_db_initially=0 --expected_values_dir=$exp --iterpercent=0 --key_len_percent_dist=1,30,69 --max_key=1000000 --max_key_len=3 --prefixpercent=0 --readpercent=0 --reopen=0 --ops_per_thread=100000000 --test_batches_snapshots=0 --value_size_mult=32 --writepercent=90 \
--use_txn=1 --txn_write_policy=0 --sync_fault_injection=1 &
pid=$!
sleep 30
kill $pid
sleep 1
```
- Run the following command to verify recovery of the crashed db under debugger. Compare the step-wise result with WAL records (e.g, WriteBatch content, xid, prepare/commit/rollback marker)
```
./db_stress \
--clear_column_family_one_in=0 --column_families=1 --db=$db --delpercent=10 --delrangepercent=0 --destroy_db_initially=0 --expected_values_dir=$exp --iterpercent=0 --key_len_percent_dist=1,30,69 --max_key=1000000 --max_key_len=3 --prefixpercent=0 --readpercent=0 --reopen=0 --ops_per_thread=100000000 --test_batches_snapshots=0 --value_size_mult=32 --writepercent=90 \
--use_txn=1 --txn_write_policy=0 --sync_fault_injection=1
```
3. Automatic testing by triggering all RocksDB stress/crash test jobs for 3 rounds with no failure.
Reviewed By: ajkr, riversand963
Differential Revision: D39199373
Pulled By: hx235
fbshipit-source-id: 7a1dec0e3e2ee6ea86ddf5dd19ceb5543a3d6f0c
Summary:
As title
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10717
Test Plan:
Unit Tests
CI
Reviewed By: riversand963
Differential Revision: D39700707
Pulled By: gitbw95
fbshipit-source-id: 54de27e695535a50159f5f6467da36aaf21bebae
Summary:
`enable_custom_split_merge` is added for enabling the custom split and merge feature, which split the compressed value into chunks so that they may better fit jemalloc bins.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10690
Test Plan:
Unit Tests
Stress Tests
Reviewed By: anand1976
Differential Revision: D39567604
Pulled By: gitbw95
fbshipit-source-id: f6d1d46200f365220055f793514601dcb0edc4b7
Summary:
This change establishes a distinctive name for the experimental new lock-free clock cache (originally developed by guidotag and revamped in PR https://github.com/facebook/rocksdb/issues/10626). A few reasons:
* We want to make it clear that this is a fundamentally different implementation vs. the old clock cache, to avoid people saying "I already tried clock cache."
* We want to highlight the key feature: it's fast (especially under parallel load)
* Because it requires an estimated charge per entry, it is not drop-in API compatible with old clock cache. This estimate might always be required for highest performance, and giving it a distinct name should reduce confusion about the distinct API requirements.
* We might develop a variant requiring the same estimate parameter but with LRU eviction. In that case, using the name HyperLRUCache should make things more clear. (FastLRUCache is just a prototype that might soon be removed.)
Some API detail:
* To reduce copy-pasting parameter lists, etc. as in LRUCache construction, I have a `MakeSharedCache()` function on `HyperClockCacheOptions` instead of `NewHyperClockCache()`.
* Changes -cache_type=clock_cache to -cache_type=hyper_clock_cache for applicable tools. I think this is more consistent / sustainable for reasons already stated.
For performance tests see https://github.com/facebook/rocksdb/pull/10626
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10684
Test Plan: no interesting functional changes; tests updated
Reviewed By: anand1976
Differential Revision: D39547800
Pulled By: pdillinger
fbshipit-source-id: 5c0fe1b5cf3cb680ab369b928c8569682b9795bf
Summary:
* Consolidates most metadata into a single word per slot so that more
can be accomplished with a single atomic update. In the common case,
Lookup was previously about 4 atomic updates, now just 1 atomic update.
Common case Release was previously 1 atomic read + 1 atomic update,
now just 1 atomic update.
* Eliminate spins / waits / yields, which likely threaten some "lock free"
benefits. Compare-exchange loops are only used in explicit Erase, and
strict_capacity_limit=true Insert. Eviction uses opportunistic compare-
exchange.
* Relaxes some aggressiveness and guarantees. For example,
* Duplicate Inserts will sometimes go undetected and the shadow duplicate
will age out with eviction.
* In many cases, the older Inserted value for a given cache key will be kept
(i.e. Insert does not support overwrite).
* Entries explicitly erased (rather than evicted) might not be freed
immediately in some rare cases.
* With strict_capacity_limit=false, capacity limit is not tracked/enforced as
precisely as LRUCache, but is self-correcting and should only deviate by a
very small number of extra or fewer entries.
* Use smaller "computed default" number of cache shards in many cases,
because benefits to larger usage tracking / eviction pools outweigh the small
cost of more lock-free atomic contention. The improvement in CPU and I/O
is dramatic in some limit-memory cases.
* Even without the sharding change, the eviction algorithm is likely more
effective than LRU overall because it's more stateful, even though the
"hot path" state tracking for it is essentially free with ref counting. It
is like a generalized CLOCK with aging (see code comments). I don't have
performance numbers showing a specific improvement, but in theory, for a
Poisson access pattern to each block, keeping some state allows better
estimation of time to next access (Poisson interval) than strict LRU. The
bounded randomness in CLOCK can also reduce "cliff" effect for repeated
range scans approaching and exceeding cache size.
## Hot path algorithm comparison
Rough descriptions, focusing on number and kind of atomic operations:
* Old `Lookup()` (2-5 atomic updates per probe):
```
Loop:
Increment internal ref count at slot
If possible hit:
Check flags atomic (and non-atomic fields)
If cache hit:
Three distinct updates to 'flags' atomic
Increment refs for internal-to-external
Return
Decrement internal ref count
while atomic read 'displacements' > 0
```
* New `Lookup()` (1-2 atomic updates per probe):
```
Loop:
Increment acquire counter in meta word (optimistic)
If visible entry (already read meta word):
If match (read non-atomic fields):
Return
Else:
Decrement acquire counter in meta word
Else if invisible entry (rare, already read meta word):
Decrement acquire counter in meta word
while atomic read 'displacements' > 0
```
* Old `Release()` (1 atomic update, conditional on atomic read, rarely more):
```
Read atomic ref count
If last reference and invisible (rare):
Use CAS etc. to remove
Return
Else:
Decrement ref count
```
* New `Release()` (1 unconditional atomic update, rarely more):
```
Increment release counter in meta word
If last reference and invisible (rare):
Use CAS etc. to remove
Return
```
## Performance test setup
Build DB with
```
TEST_TMPDIR=/dev/shm ./db_bench -benchmarks=fillrandom -num=30000000 -disable_wal=1 -bloom_bits=16
```
Test with
```
TEST_TMPDIR=/dev/shm ./db_bench -benchmarks=readrandom -readonly -num=30000000 -bloom_bits=16 -cache_index_and_filter_blocks=1 -cache_size=${CACHE_MB}000000 -duration 60 -threads=$THREADS -statistics
```
Numbers on a single socket Skylake Xeon system with 48 hardware threads, DEBUG_LEVEL=0 PORTABLE=0. Very similar story on a dual socket system with 80 hardware threads. Using (every 2nd) Fibonacci MB cache sizes to sample the territory between powers of two. Configurations:
base: LRUCache before this change, but with db_bench change to default cache_numshardbits=-1 (instead of fixed at 6)
folly: LRUCache before this change, with folly enabled (distributed mutex) but on an old compiler (sorry)
gt_clock: experimental ClockCache before this change
new_clock: experimental ClockCache with this change
## Performance test results
First test "hot path" read performance, with block cache large enough for whole DB:
4181MB 1thread base -> kops/s: 47.761
4181MB 1thread folly -> kops/s: 45.877
4181MB 1thread gt_clock -> kops/s: 51.092
4181MB 1thread new_clock -> kops/s: 53.944
4181MB 16thread base -> kops/s: 284.567
4181MB 16thread folly -> kops/s: 249.015
4181MB 16thread gt_clock -> kops/s: 743.762
4181MB 16thread new_clock -> kops/s: 861.821
4181MB 24thread base -> kops/s: 303.415
4181MB 24thread folly -> kops/s: 266.548
4181MB 24thread gt_clock -> kops/s: 975.706
4181MB 24thread new_clock -> kops/s: 1205.64 (~= 24 * 53.944)
4181MB 32thread base -> kops/s: 311.251
4181MB 32thread folly -> kops/s: 274.952
4181MB 32thread gt_clock -> kops/s: 1045.98
4181MB 32thread new_clock -> kops/s: 1370.38
4181MB 48thread base -> kops/s: 310.504
4181MB 48thread folly -> kops/s: 268.322
4181MB 48thread gt_clock -> kops/s: 1195.65
4181MB 48thread new_clock -> kops/s: 1604.85 (~= 24 * 1.25 * 53.944)
4181MB 64thread base -> kops/s: 307.839
4181MB 64thread folly -> kops/s: 272.172
4181MB 64thread gt_clock -> kops/s: 1204.47
4181MB 64thread new_clock -> kops/s: 1615.37
4181MB 128thread base -> kops/s: 310.934
4181MB 128thread folly -> kops/s: 267.468
4181MB 128thread gt_clock -> kops/s: 1188.75
4181MB 128thread new_clock -> kops/s: 1595.46
Whether we have just one thread on a quiet system or an overload of threads, the new version wins every time in thousand-ops per second, sometimes dramatically so. Mutex-based implementation quickly becomes contention-limited. New clock cache shows essentially perfect scaling up to number of physical cores (24), and then each hyperthreaded core adding about 1/4 the throughput of an additional physical core (see 48 thread case). Block cache miss rates (omitted above) are negligible across the board. With partitioned instead of full filters, the maximum speed-up vs. base is more like 2.5x rather than 5x.
Now test a large block cache with low miss ratio, but some eviction is required:
1597MB 1thread base -> kops/s: 46.603 io_bytes/op: 1584.63 miss_ratio: 0.0201066 max_rss_mb: 1589.23
1597MB 1thread folly -> kops/s: 45.079 io_bytes/op: 1530.03 miss_ratio: 0.019872 max_rss_mb: 1550.43
1597MB 1thread gt_clock -> kops/s: 48.711 io_bytes/op: 1566.63 miss_ratio: 0.0198923 max_rss_mb: 1691.4
1597MB 1thread new_clock -> kops/s: 51.531 io_bytes/op: 1589.07 miss_ratio: 0.0201969 max_rss_mb: 1583.56
1597MB 32thread base -> kops/s: 301.174 io_bytes/op: 1439.52 miss_ratio: 0.0184218 max_rss_mb: 1656.59
1597MB 32thread folly -> kops/s: 273.09 io_bytes/op: 1375.12 miss_ratio: 0.0180002 max_rss_mb: 1586.8
1597MB 32thread gt_clock -> kops/s: 904.497 io_bytes/op: 1411.29 miss_ratio: 0.0179934 max_rss_mb: 1775.89
1597MB 32thread new_clock -> kops/s: 1182.59 io_bytes/op: 1440.77 miss_ratio: 0.0185449 max_rss_mb: 1636.45
1597MB 128thread base -> kops/s: 309.91 io_bytes/op: 1438.25 miss_ratio: 0.018399 max_rss_mb: 1689.98
1597MB 128thread folly -> kops/s: 267.605 io_bytes/op: 1394.16 miss_ratio: 0.0180286 max_rss_mb: 1631.91
1597MB 128thread gt_clock -> kops/s: 691.518 io_bytes/op: 9056.73 miss_ratio: 0.0186572 max_rss_mb: 1982.26
1597MB 128thread new_clock -> kops/s: 1406.12 io_bytes/op: 1440.82 miss_ratio: 0.0185463 max_rss_mb: 1685.63
610MB 1thread base -> kops/s: 45.511 io_bytes/op: 2279.61 miss_ratio: 0.0290528 max_rss_mb: 615.137
610MB 1thread folly -> kops/s: 43.386 io_bytes/op: 2217.29 miss_ratio: 0.0289282 max_rss_mb: 600.996
610MB 1thread gt_clock -> kops/s: 46.207 io_bytes/op: 2275.51 miss_ratio: 0.0290057 max_rss_mb: 637.934
610MB 1thread new_clock -> kops/s: 48.879 io_bytes/op: 2283.1 miss_ratio: 0.0291253 max_rss_mb: 613.5
610MB 32thread base -> kops/s: 306.59 io_bytes/op: 2250 miss_ratio: 0.0288721 max_rss_mb: 683.402
610MB 32thread folly -> kops/s: 269.176 io_bytes/op: 2187.86 miss_ratio: 0.0286938 max_rss_mb: 628.742
610MB 32thread gt_clock -> kops/s: 855.097 io_bytes/op: 2279.26 miss_ratio: 0.0288009 max_rss_mb: 733.062
610MB 32thread new_clock -> kops/s: 1121.47 io_bytes/op: 2244.29 miss_ratio: 0.0289046 max_rss_mb: 666.453
610MB 128thread base -> kops/s: 305.079 io_bytes/op: 2252.43 miss_ratio: 0.0288884 max_rss_mb: 723.457
610MB 128thread folly -> kops/s: 269.583 io_bytes/op: 2204.58 miss_ratio: 0.0287001 max_rss_mb: 676.426
610MB 128thread gt_clock -> kops/s: 53.298 io_bytes/op: 8128.98 miss_ratio: 0.0292452 max_rss_mb: 956.273
610MB 128thread new_clock -> kops/s: 1301.09 io_bytes/op: 2246.04 miss_ratio: 0.0289171 max_rss_mb: 788.812
The new version is still winning every time, sometimes dramatically so, and we can tell from the maximum resident memory numbers (which contain some noise, by the way) that the new cache is not cheating on memory usage. IMPORTANT: The previous generation experimental clock cache appears to hit a serious bottleneck in the higher thread count configurations, presumably due to some of its waiting functionality. (The same bottleneck is not seen with partitioned index+filters.)
Now we consider even smaller cache sizes, with higher miss ratios, eviction work, etc.
233MB 1thread base -> kops/s: 10.557 io_bytes/op: 227040 miss_ratio: 0.0403105 max_rss_mb: 247.371
233MB 1thread folly -> kops/s: 15.348 io_bytes/op: 112007 miss_ratio: 0.0372238 max_rss_mb: 245.293
233MB 1thread gt_clock -> kops/s: 6.365 io_bytes/op: 244854 miss_ratio: 0.0413873 max_rss_mb: 259.844
233MB 1thread new_clock -> kops/s: 47.501 io_bytes/op: 2591.93 miss_ratio: 0.0330989 max_rss_mb: 242.461
233MB 32thread base -> kops/s: 96.498 io_bytes/op: 363379 miss_ratio: 0.0459966 max_rss_mb: 479.227
233MB 32thread folly -> kops/s: 109.95 io_bytes/op: 314799 miss_ratio: 0.0450032 max_rss_mb: 400.738
233MB 32thread gt_clock -> kops/s: 2.353 io_bytes/op: 385397 miss_ratio: 0.048445 max_rss_mb: 500.688
233MB 32thread new_clock -> kops/s: 1088.95 io_bytes/op: 2567.02 miss_ratio: 0.0330593 max_rss_mb: 303.402
233MB 128thread base -> kops/s: 84.302 io_bytes/op: 378020 miss_ratio: 0.0466558 max_rss_mb: 1051.84
233MB 128thread folly -> kops/s: 89.921 io_bytes/op: 338242 miss_ratio: 0.0460309 max_rss_mb: 812.785
233MB 128thread gt_clock -> kops/s: 2.588 io_bytes/op: 462833 miss_ratio: 0.0509158 max_rss_mb: 1109.94
233MB 128thread new_clock -> kops/s: 1299.26 io_bytes/op: 2565.94 miss_ratio: 0.0330531 max_rss_mb: 361.016
89MB 1thread base -> kops/s: 0.574 io_bytes/op: 5.35977e+06 miss_ratio: 0.274427 max_rss_mb: 91.3086
89MB 1thread folly -> kops/s: 0.578 io_bytes/op: 5.16549e+06 miss_ratio: 0.27276 max_rss_mb: 96.8984
89MB 1thread gt_clock -> kops/s: 0.512 io_bytes/op: 4.13111e+06 miss_ratio: 0.242817 max_rss_mb: 119.441
89MB 1thread new_clock -> kops/s: 48.172 io_bytes/op: 2709.76 miss_ratio: 0.0346162 max_rss_mb: 100.754
89MB 32thread base -> kops/s: 5.779 io_bytes/op: 6.14192e+06 miss_ratio: 0.320399 max_rss_mb: 311.812
89MB 32thread folly -> kops/s: 5.601 io_bytes/op: 5.83838e+06 miss_ratio: 0.313123 max_rss_mb: 252.418
89MB 32thread gt_clock -> kops/s: 0.77 io_bytes/op: 3.99236e+06 miss_ratio: 0.236296 max_rss_mb: 396.422
89MB 32thread new_clock -> kops/s: 1064.97 io_bytes/op: 2687.23 miss_ratio: 0.0346134 max_rss_mb: 155.293
89MB 128thread base -> kops/s: 4.959 io_bytes/op: 6.20297e+06 miss_ratio: 0.323945 max_rss_mb: 823.43
89MB 128thread folly -> kops/s: 4.962 io_bytes/op: 5.9601e+06 miss_ratio: 0.319857 max_rss_mb: 626.824
89MB 128thread gt_clock -> kops/s: 1.009 io_bytes/op: 4.1083e+06 miss_ratio: 0.242512 max_rss_mb: 1095.32
89MB 128thread new_clock -> kops/s: 1224.39 io_bytes/op: 2688.2 miss_ratio: 0.0346207 max_rss_mb: 218.223
^ Now something interesting has happened: the new clock cache has gained a dramatic lead in the single-threaded case, and this is because the cache is so small, and full filters are so big, that dividing the cache into 64 shards leads to significant (random) imbalances in cache shards and excessive churn in imbalanced shards. This new clock cache only uses two shards for this configuration, and that helps to ensure that entries are part of a sufficiently big pool that their eviction order resembles the single-shard order. (This effect is not seen with partitioned index+filters.)
Even smaller cache size:
34MB 1thread base -> kops/s: 0.198 io_bytes/op: 1.65342e+07 miss_ratio: 0.939466 max_rss_mb: 48.6914
34MB 1thread folly -> kops/s: 0.201 io_bytes/op: 1.63416e+07 miss_ratio: 0.939081 max_rss_mb: 45.3281
34MB 1thread gt_clock -> kops/s: 0.448 io_bytes/op: 4.43957e+06 miss_ratio: 0.266749 max_rss_mb: 100.523
34MB 1thread new_clock -> kops/s: 1.055 io_bytes/op: 1.85439e+06 miss_ratio: 0.107512 max_rss_mb: 75.3125
34MB 32thread base -> kops/s: 3.346 io_bytes/op: 1.64852e+07 miss_ratio: 0.93596 max_rss_mb: 180.48
34MB 32thread folly -> kops/s: 3.431 io_bytes/op: 1.62857e+07 miss_ratio: 0.935693 max_rss_mb: 137.531
34MB 32thread gt_clock -> kops/s: 1.47 io_bytes/op: 4.89704e+06 miss_ratio: 0.295081 max_rss_mb: 392.465
34MB 32thread new_clock -> kops/s: 8.19 io_bytes/op: 3.70456e+06 miss_ratio: 0.20826 max_rss_mb: 519.793
34MB 128thread base -> kops/s: 2.293 io_bytes/op: 1.64351e+07 miss_ratio: 0.931866 max_rss_mb: 449.484
34MB 128thread folly -> kops/s: 2.34 io_bytes/op: 1.6219e+07 miss_ratio: 0.932023 max_rss_mb: 396.457
34MB 128thread gt_clock -> kops/s: 1.798 io_bytes/op: 5.4241e+06 miss_ratio: 0.324881 max_rss_mb: 1104.41
34MB 128thread new_clock -> kops/s: 10.519 io_bytes/op: 2.39354e+06 miss_ratio: 0.136147 max_rss_mb: 1050.52
As the miss ratio gets higher (say, above 10%), the CPU time spent in eviction starts to erode the advantage of using fewer shards (13% miss rate much lower than 94%). LRU's O(1) eviction time can eventually pay off when there's enough block cache churn:
13MB 1thread base -> kops/s: 0.195 io_bytes/op: 1.65732e+07 miss_ratio: 0.946604 max_rss_mb: 45.6328
13MB 1thread folly -> kops/s: 0.197 io_bytes/op: 1.63793e+07 miss_ratio: 0.94661 max_rss_mb: 33.8633
13MB 1thread gt_clock -> kops/s: 0.519 io_bytes/op: 4.43316e+06 miss_ratio: 0.269379 max_rss_mb: 100.684
13MB 1thread new_clock -> kops/s: 0.176 io_bytes/op: 1.54148e+07 miss_ratio: 0.91545 max_rss_mb: 66.2383
13MB 32thread base -> kops/s: 3.266 io_bytes/op: 1.65544e+07 miss_ratio: 0.943386 max_rss_mb: 132.492
13MB 32thread folly -> kops/s: 3.396 io_bytes/op: 1.63142e+07 miss_ratio: 0.943243 max_rss_mb: 101.863
13MB 32thread gt_clock -> kops/s: 2.758 io_bytes/op: 5.13714e+06 miss_ratio: 0.310652 max_rss_mb: 396.121
13MB 32thread new_clock -> kops/s: 3.11 io_bytes/op: 1.23419e+07 miss_ratio: 0.708425 max_rss_mb: 321.758
13MB 128thread base -> kops/s: 2.31 io_bytes/op: 1.64823e+07 miss_ratio: 0.939543 max_rss_mb: 425.539
13MB 128thread folly -> kops/s: 2.339 io_bytes/op: 1.6242e+07 miss_ratio: 0.939966 max_rss_mb: 346.098
13MB 128thread gt_clock -> kops/s: 3.223 io_bytes/op: 5.76928e+06 miss_ratio: 0.345899 max_rss_mb: 1087.77
13MB 128thread new_clock -> kops/s: 2.984 io_bytes/op: 1.05341e+07 miss_ratio: 0.606198 max_rss_mb: 898.27
gt_clock is clearly blowing way past its memory budget for lower miss rates and best throughput. new_clock also seems to be exceeding budgets, and this warrants more investigation but is not the use case we are targeting with the new cache. With partitioned index+filter, the miss ratio is much better, and although still high enough that the eviction CPU time is definitely offsetting mutex contention:
13MB 1thread base -> kops/s: 16.326 io_bytes/op: 23743.9 miss_ratio: 0.205362 max_rss_mb: 65.2852
13MB 1thread folly -> kops/s: 15.574 io_bytes/op: 19415 miss_ratio: 0.184157 max_rss_mb: 56.3516
13MB 1thread gt_clock -> kops/s: 14.459 io_bytes/op: 22873 miss_ratio: 0.198355 max_rss_mb: 63.9688
13MB 1thread new_clock -> kops/s: 16.34 io_bytes/op: 24386.5 miss_ratio: 0.210512 max_rss_mb: 61.707
13MB 128thread base -> kops/s: 289.786 io_bytes/op: 23710.9 miss_ratio: 0.205056 max_rss_mb: 103.57
13MB 128thread folly -> kops/s: 185.282 io_bytes/op: 19433.1 miss_ratio: 0.184275 max_rss_mb: 116.219
13MB 128thread gt_clock -> kops/s: 354.451 io_bytes/op: 23150.6 miss_ratio: 0.200495 max_rss_mb: 102.871
13MB 128thread new_clock -> kops/s: 295.359 io_bytes/op: 24626.4 miss_ratio: 0.212452 max_rss_mb: 121.109
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10626
Test Plan: updated unit tests, stress/crash test runs including with TSAN, ASAN, UBSAN
Reviewed By: anand1976
Differential Revision: D39368406
Pulled By: pdillinger
fbshipit-source-id: 5afc44da4c656f8f751b44552bbf27bd3ca6fef9
Summary:
Hopefully, we can re-enable the combination of user-defined timestamp and subcompactions
after https://github.com/facebook/rocksdb/issues/10658.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10689
Test Plan:
Make sure the following succeeds on devserver.
make crash_test_with_ts
Reviewed By: ltamasi
Differential Revision: D39556558
Pulled By: riversand963
fbshipit-source-id: 4695f420b1bc9ebf3b24640b693746f4db82c149
Summary:
The patch makes it possible to use the `JemallocNodumpAllocator` with the
block/blob caches in `db_bench`. In addition to its stated purpose of excluding
cache contents from core dumps, `JemallocNodumpAllocator` also uses
a dedicated arena and jemalloc tcaches for cache allocations, which can
reduce fragmentation and thus memory usage.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10685
Reviewed By: riversand963
Differential Revision: D39552261
Pulled By: ltamasi
fbshipit-source-id: b5c58eab6b7c1baa9a307d9f1248df1d7a77d2b5
Summary:
Same as title
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10632
Test Plan: make crash_test -j32
Reviewed By: anand1976
Differential Revision: D39241479
Pulled By: akankshamahajan15
fbshipit-source-id: 5db5b0c007da786bacc1b30d8926d36d6d029b87
Summary:
Expanded `all_params` to include all parameters crash test may set. Previously, `atomic_flush` was not included in `all_params` and thus was not visible to `finalize_and_sanitize()`. The consequence was manual crash test runs could provide unsafe combinations of parameters to `db_stress`. For example, running `db_crashtest.py` with `-atomic_flush=0` could cause `db_stress` to run with `-atomic_flush=0 -disable_wal=1`, which is known to produce inconsistencies across column families.
While expanding `all_params`, I found we cannot have an entry in it for both `db_stress` and `db_crashtest.py`. So I renamed `enable_tiered_storage` to `test_tiered_storage` for `db_crashtest.py`, which appears more conventional anyways.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10654
Reviewed By: hx235
Differential Revision: D39369349
Pulled By: ajkr
fbshipit-source-id: 31d9010c760c868b20d5e9bd78ba75c8ff3ce348
Summary:
**Context/Summary:**
`ExpectedState` is not aware of transaction-related concept so `use_txn=1 ` is not compatible with `sync_fault_injection=1`. Therefore this PR disabled this combination until we expand our correctness testing to transaction related features.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10605
Test Plan:
- Run the following commands to verify `--use_txn` is correctly sanitized
- `python3 ./tools/db_crashtest.py blackbox --use_txn=1 --sync_fault_injection=1 `
- `python3 ./tools/db_crashtest.py blackbox --use_txn=0 --sync_fault_injection=1 `
Reviewed By: ajkr
Differential Revision: D39121287
Pulled By: hx235
fbshipit-source-id: 7d5d6dd32479ea1c07df4f38322650f3a60def9c
Summary:
Currently, `db_bench` and `db_stress` print the blob cache options even if
a shared block/blob cache is configured, i.e. when they are not actually
in effect. The patch changes this so they are only printed when a separate blob
cache is used.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10614
Test Plan: Tested manually using `db_bench` and `db_stress`.
Reviewed By: akankshamahajan15
Differential Revision: D39144603
Pulled By: ltamasi
fbshipit-source-id: f714304c5d46186f8514746c27ee6f52aa3e4af8
Summary:
**Context:**
Below crash test revealed a bug that directory containing CURRENT file (short for `dir_contains_current_file` below) was not always get synced after a new CURRENT is created and being called with `RenameFile` as part of the creation.
This bug exposes a risk that such un-synced directory containing the updated CURRENT can’t survive a host crash (e.g, power loss) hence get corrupted. This then will be followed by a recovery from a corrupted CURRENT that we don't want.
The root-cause is that a nullptr `FSDirectory* dir_contains_current_file` sometimes gets passed-down to `SetCurrentFile()` hence in those case `dir_contains_current_file->FSDirectory::FsyncWithDirOptions()` will be skipped (which otherwise will internally call`Env/FS::SyncDic()` )
```
./db_stress --acquire_snapshot_one_in=10000 --adaptive_readahead=1 --allow_data_in_errors=True --avoid_unnecessary_blocking_io=0 --backup_max_size=104857600 --backup_one_in=100000 --batch_protection_bytes_per_key=8 --block_size=16384 --bloom_bits=134.8015470676662 --bottommost_compression_type=disable --cache_size=8388608 --checkpoint_one_in=1000000 --checksum_type=kCRC32c --clear_column_family_one_in=0 --compact_files_one_in=1000000 --compact_range_one_in=1000000 --compaction_pri=2 --compaction_ttl=100 --compression_max_dict_buffer_bytes=511 --compression_max_dict_bytes=16384 --compression_type=zstd --compression_use_zstd_dict_trainer=1 --compression_zstd_max_train_bytes=65536 --continuous_verification_interval=0 --data_block_index_type=0 --db=$db --db_write_buffer_size=1048576 --delpercent=5 --delrangepercent=0 --destroy_db_initially=0 --disable_wal=0 --enable_compaction_filter=0 --enable_pipelined_write=1 --expected_values_dir=$exp --fail_if_options_file_error=1 --file_checksum_impl=none --flush_one_in=1000000 --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=4 --ingest_external_file_one_in=0 --iterpercent=10 --key_len_percent_dist=1,30,69 --level_compaction_dynamic_level_bytes=True --mark_for_compaction_one_file_in=10 --max_background_compactions=20 --max_bytes_for_level_base=10485760 --max_key=10000 --max_key_len=3 --max_manifest_file_size=16384 --max_write_batch_group_size_bytes=64 --max_write_buffer_number=3 --max_write_buffer_size_to_maintain=0 --memtable_prefix_bloom_size_ratio=0.001 --memtable_protection_bytes_per_key=1 --memtable_whole_key_filtering=1 --mmap_read=1 --nooverwritepercent=1 --open_metadata_write_fault_one_in=0 --open_read_fault_one_in=0 --open_write_fault_one_in=0 --ops_per_thread=100000000 --optimize_filters_for_memory=1 --paranoid_file_checks=1 --partition_pinning=2 --pause_background_one_in=1000000 --periodic_compaction_seconds=0 --prefix_size=5 --prefixpercent=5 --prepopulate_block_cache=1 --progress_reports=0 --read_fault_one_in=1000 --readpercent=45 --recycle_log_file_num=0 --reopen=0 --ribbon_starting_level=999 --secondary_cache_fault_one_in=32 --secondary_cache_uri=compressed_secondary_cache://capacity=8388608 --set_options_one_in=10000 --snapshot_hold_ops=100000 --sst_file_manager_bytes_per_sec=0 --sst_file_manager_bytes_per_truncate=0 --subcompactions=3 --sync_fault_injection=1 --target_file_size_base=2097 --target_file_size_multiplier=2 --test_batches_snapshots=1 --top_level_index_pinning=1 --use_full_merge_v1=1 --use_merge=1 --value_size_mult=32 --verify_checksum=1 --verify_checksum_one_in=1000000 --verify_db_one_in=100000 --verify_sst_unique_id_in_manifest=1 --wal_bytes_per_sync=524288 --write_buffer_size=4194 --writepercent=35
```
```
stderr:
WARNING: prefix_size is non-zero but memtablerep != prefix_hash
db_stress: utilities/fault_injection_fs.cc:748: virtual rocksdb::IOStatus rocksdb::FaultInjectionTestFS::RenameFile(const std::string &, const std::string &, const rocksdb::IOOptions &, rocksdb::IODebugContext *): Assertion `tlist.find(tdn.second) == tlist.end()' failed.`
```
**Summary:**
The PR ensured the non-test path pass down a non-null dir containing CURRENT (which is by current RocksDB assumption just db_dir) by doing the following:
- Renamed `directory_to_fsync` as `dir_contains_current_file` in `SetCurrentFile()` to tighten the association between this directory and CURRENT file
- Changed `SetCurrentFile()` API to require `dir_contains_current_file` being passed-in, instead of making it by default nullptr.
- Because `SetCurrentFile()`'s `dir_contains_current_file` is passed down from `VersionSet::LogAndApply()` then `VersionSet::ProcessManifestWrites()` (i.e, think about this as a chain of 3 functions related to MANIFEST update), these 2 functions also got refactored to require `dir_contains_current_file`
- Updated the non-test-path callers of these 3 functions to obtain and pass in non-nullptr `dir_contains_current_file`, which by current assumption of RocksDB, is the `FSDirectory* db_dir`.
- `db_impl` path will obtain `DBImpl::directories_.getDbDir()` while others with no access to such `directories_` are obtained on the fly by creating such object `FileSystem::NewDirectory(..)` and manage it by unique pointers to ensure short life time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10573
Test Plan:
- `make check`
- Passed the repro db_stress command
- For future improvement, since we currently don't assert dir containing CURRENT to be non-nullptr due to https://github.com/facebook/rocksdb/pull/10573#pullrequestreview-1087698899, there is still chances that future developers mistakenly pass down nullptr dir containing CURRENT thus resulting skipped sync dir and cause the bug again. Therefore a smarter test (e.g, such as quoted from ajkr "(make) unsynced data loss to be dropping files corresponding to unsynced directory entries") is still needed.
Reviewed By: ajkr
Differential Revision: D39005886
Pulled By: hx235
fbshipit-source-id: 336fb9090d0cfa6ca3dd580db86268007dde7f5a
Summary:
CI benchmarks refine configuration
Run only “essential” benchmarks, but for longer
Fix (reduce) the NUM_KEYS to ensure cached behaviour
Reduce level size to try to ensure more levels
Refine test durations again, more time per test, but fewer tests.
In CI benchmark mode, the only read test is readrandom.
There are still 3 mostly-read tests.
Goal is to squeeze complete run a little bit inside 1 hour so it doesn’t clash with the next run (cron scheduled for main branch), but it gets to run as long as possible, so that results are as credible as possible.
Reduce thread count to physical capacity, in an attempt to reduce throughput variance for write heavy tests. See Mark Callaghan’s comments in related documentation..
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10514
Reviewed By: ajkr
Differential Revision: D38952469
Pulled By: jay-zhuang
fbshipit-source-id: 72fa6bba897cc47066ced65facd1fd36e28f30a8
Summary:
The features that cannot work with disable_wal=1 due to unsynced data dropping (ingest_external_file_one_in and enable_compaction_filter) similarly cannot work with sync_fault_injection=1. This PR prevents those features from being used together with sync_fault_injection=1.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10559
Reviewed By: hx235
Differential Revision: D38953019
Pulled By: ajkr
fbshipit-source-id: 7e2c7644ec84d7323f632cf976bcee00502d0ed7
Summary:
As mentioned in https://github.com/facebook/rocksdb/pull/5506#issuecomment-506021913,
`db_stress` does not have much verification for iterator correctness.
It has a `TestIterate()` function, but that is mainly for comparing results
between two iterators, one with `total_order_seek` and the other optionally
sets auto_prefix, upper/lower bounds. Commit 49a0581ad2462e31aa3f768afa769e0d33390f33
added a new `TestIterateAgainstExpected()` function that compares iterator against
expected state. It locks a range of keys, creates an iterator, does
a random sequence of `Next/Prev` and compares against expected state.
This PR is based on that commit, the main changes include some logs
(for easier debugging if a test fails), a forward and backward scan to
cover the entire locked key range, and a flag for optionally turning on
this version of Iterator testing.
Added constraint that the checks against expected state in
`TestIterateAgainstExpected()` and in `TestGet()` are only turned on
when `--skip_verifydb` flag is not set.
Remove the change log introduced in https://github.com/facebook/rocksdb/issues/10553.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10538
Test Plan:
Run `db_stress` with `--verify_iterator_with_expected_state_one_in=1`,
and a large `--iterpercent` and `--num_iterations`. Checked `op_logs`
manually to ensure expected coverage. Tweaked part of the code in
https://github.com/facebook/rocksdb/issues/10449 and stress test was able to catch it.
- internally run various flavor of crash test
Reviewed By: ajkr
Differential Revision: D38847269
Pulled By: cbi42
fbshipit-source-id: 8b4402a9bba9f6cfa08051943cd672579d489599
Summary:
Optionally issue DeleteRange in `*whilewriting` benchmarks. This happens in `BGWriter` and uses similar logic as in `DoWrite` to issue DeleteRange operations. I added this when I was benchmarking https://github.com/facebook/rocksdb/issues/10547, but this should be an independent PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10552
Test Plan: ran some benchmarks with various delete range options, e.g. `./db_bench --benchmarks=readwhilewriting --writes_per_range_tombstone=100 --writes=200000 --reads=1000000 --disable_auto_compactions --max_num_range_tombstones=10000`
Reviewed By: ajkr
Differential Revision: D38927020
Pulled By: cbi42
fbshipit-source-id: 31ee20cb8127f7173f0816ea0cc2a204ec02aad6
Summary:
After branch 7.6.fb branch is cut, following release process, upgrade version number to 7.7 and add 7.6.fb to format compatibility check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10546
Test Plan: Watch CI
Reviewed By: ajkr
Differential Revision: D38892023
Pulled By: gitbw95
fbshipit-source-id: 94e96dedbd973f5f9713e73d3bed336e4678565b
Summary:
This PR exploits parallelism in MultiGet across levels. It applies only to the coroutine version of MultiGet. Previously, MultiGet file reads from SST files in the same level were parallelized. With this PR, MultiGet batches with keys distributed across multiple levels are read in parallel. This is accomplished by splitting the keys not present in a level (determined by bloom filtering) into a separate batch, and processing the new batch in parallel with the original batch.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10535
Test Plan:
1. Ensure existing MultiGet unit tests pass, updating them as necessary
2. New unit tests - TODO
3. Run stress test - TODO
No noticeable regression (<1%) without async IO -
Without PR: `multireadrandom : 7.261 micros/op 1101724 ops/sec 60.007 seconds 66110936 operations; 571.6 MB/s (8168992 of 8168992 found)`
With PR: `multireadrandom : 7.305 micros/op 1095167 ops/sec 60.007 seconds 65717936 operations; 568.2 MB/s (8271992 of 8271992 found)`
For a fully cached DB, but with async IO option on, no regression observed (<1%) -
Without PR: `multireadrandom : 5.201 micros/op 1538027 ops/sec 60.005 seconds 92288936 operations; 797.9 MB/s (11540992 of 11540992 found) `
With PR: `multireadrandom : 5.249 micros/op 1524097 ops/sec 60.005 seconds 91452936 operations; 790.7 MB/s (11649992 of 11649992 found) `
Reviewed By: akankshamahajan15
Differential Revision: D38774009
Pulled By: anand1976
fbshipit-source-id: c955e259749f1c091590ade73105b3ee46cd0007
Summary:
RocksDB's `Cache` abstraction currently supports two priority levels for items: high (used for frequently accessed/highly valuable SST metablocks like index/filter blocks) and low (used for SST data blocks). Blobs are typically lower-value targets for caching than data blocks, since 1) with BlobDB, data blocks containing blob references conceptually form an index structure which has to be consulted before we can read the blob value, and 2) cached blobs represent only a single key-value, while cached data blocks generally contain multiple KVs. Since we would like to make it possible to use the same backing cache for the block cache and the blob cache, it would make sense to add a new, lower-than-low cache priority level (bottom level) for blobs so data blocks are prioritized over them.
This task is a part of https://github.com/facebook/rocksdb/issues/10156
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10461
Reviewed By: siying
Differential Revision: D38672823
Pulled By: ltamasi
fbshipit-source-id: 90cf7362036563d79891f47be2cc24b827482743
Summary:
Some files miss headers. Also some headers are irregular. Fix them to make an internal checkup tool happy.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10519
Reviewed By: jay-zhuang
Differential Revision: D38603291
fbshipit-source-id: 13b1bbd6d48f5ee15ba20da67544396de48238f1
Summary:
Moved linux builds to using docker to avoid CI instability caused by dependency installation site down.
Added the `Dockerfile` which is used to build the image.
The build time is also significantly reduced, because no dependencies installation and with using 2xlarge+ instance for slow build (like tsan test).
Also fixed a few issues detected while building this:
* `DestoryDB()` Status not checked for a few tests
* nullptr might be used in `inlineskiplist.cc`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10496
Test Plan: CI
Reviewed By: ajkr
Differential Revision: D38554200
Pulled By: jay-zhuang
fbshipit-source-id: 16e8fb2bf07b9c84bb27fb18421c4d54f2f248fd
Summary:
Change tiered compaction feature from `bottommost_temperture` to
`last_level_temperture`. The old option is kept for migration purpose only,
which is behaving the same as `last_level_temperture` and it will be removed in
the next release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10471
Test Plan: CI
Reviewed By: siying
Differential Revision: D38450621
Pulled By: jay-zhuang
fbshipit-source-id: cc1cdf8bad409376fec0152abc0a64fb72a91527
Summary:
Currently user_defined_timestamp is failing in stress test with
subcompactions. So disabling it for now and will re enable it once its
fixed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10503
Test Plan: make crash_test_with_ts -j32
Reviewed By: riversand963
Differential Revision: D38510485
Pulled By: akankshamahajan15
fbshipit-source-id: 82fd0ec8cf86a96ff6653edd5bad7623cb9e0a15
Summary:
- Right now each read fragments the memtable range tombstones https://github.com/facebook/rocksdb/issues/4808. This PR explores the idea of fragmenting memtable range tombstones in the write path and reads can just read this cached fragmented tombstone without any fragmenting cost. This PR only does the caching for immutable memtable, and does so right before a memtable is added to an immutable memtable list. The fragmentation is done without holding mutex to minimize its performance impact.
- db_bench is updated to print out the number of range deletions executed if there is any.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10380
Test Plan:
- CI, added asserts in various places to check whether a fragmented range tombstone list should have been constructed.
- Benchmark: as this PR only optimizes immutable memtable path, the number of writes in the benchmark is chosen such an immutable memtable is created and range tombstones are in that memtable.
```
single thread:
./db_bench --benchmarks=fillrandom,readrandom --writes_per_range_tombstone=1 --max_write_buffer_number=100 --min_write_buffer_number_to_merge=100 --writes=500000 --reads=100000 --max_num_range_tombstones=100
multi_thread
./db_bench --benchmarks=fillrandom,readrandom --writes_per_range_tombstone=1 --max_write_buffer_number=100 --min_write_buffer_number_to_merge=100 --writes=15000 --reads=20000 --threads=32 --max_num_range_tombstones=100
```
Commit 99cdf16464 is included in benchmark result. It was an earlier attempt where tombstones are fragmented for each write operation. Reader threads share it using a shared_ptr which would slow down multi-thread read performance as seen in benchmark results.
Results are averaged over 5 runs.
Single thread result:
| Max # tombstones | main fillrandom micros/op | 99cdf16464 | Post PR | main readrandom micros/op | 99cdf16464 | Post PR |
| ------------- | ------------- |------------- |------------- |------------- |------------- |------------- |
| 0 |6.68 |6.57 |6.72 |4.72 |4.79 |4.54 |
| 1 |6.67 |6.58 |6.62 |5.41 |4.74 |4.72 |
| 10 |6.59 |6.5 |6.56 |7.83 |4.69 |4.59 |
| 100 |6.62 |6.75 |6.58 |29.57 |5.04 |5.09 |
| 1000 |6.54 |6.82 |6.61 |320.33 |5.22 |5.21 |
32-thread result: note that "Max # tombstones" is per thread.
| Max # tombstones | main fillrandom micros/op | 99cdf16464 | Post PR | main readrandom micros/op | 99cdf16464 | Post PR |
| ------------- | ------------- |------------- |------------- |------------- |------------- |------------- |
| 0 |234.52 |260.25 |239.42 |5.06 |5.38 |5.09 |
| 1 |236.46 |262.0 |231.1 |19.57 |22.14 |5.45 |
| 10 |236.95 |263.84 |251.49 |151.73 |21.61 |5.73 |
| 100 |268.16 |296.8 |280.13 |2308.52 |22.27 |6.57 |
Reviewed By: ajkr
Differential Revision: D37916564
Pulled By: cbi42
fbshipit-source-id: 05d6d2e16df26c374c57ddcca13a5bfe9d5b731e
Summary:
This PR avoids allocations and copies for the result of `GetMergeOperands()` when the average operand size is at least 256 bytes and the total operands size is at least 32KB. The `GetMergeOperands()` already included `PinnableSlice` but was calling `PinSelf()` (i.e., allocating and copying) for each operand. When this optimization takes effect, we instead call `PinSlice()` to skip that allocation and copy. Resources are pinned in order for the `PinnableSlice` to point to valid memory even after `GetMergeOperands()` returns.
The pinned resources include a referenced `SuperVersion`, a `MergingContext`, and a `PinnedIteratorsManager`. They are bundled into a `GetMergeOperandsState`. We use `SharedCleanablePtr` to share that bundle among all `PinnableSlice`s populated by `GetMergeOperands()`. That way, the last `PinnableSlice` to be `Reset()` will cleanup the bundle, including unreferencing the `SuperVersion`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10458
Test Plan:
- new DB level test
- measured benefit/regression in a number of memtable scenarios
Setup command:
```
$ ./db_bench -benchmarks=mergerandom -merge_operator=StringAppendOperator -num=$num -writes=16384 -key_size=16 -value_size=$value_sz -compression_type=none -write_buffer_size=1048576000
```
Benchmark command:
```
./db_bench -threads=$threads -use_existing_db=true -avoid_flush_during_recovery=true -write_buffer_size=1048576000 -benchmarks=readrandomoperands -merge_operator=StringAppendOperator -num=$num -duration=10
```
Worst regression is when a key has many tiny operands:
- Parameters: num=1 (implying 16384 operands per key), value_sz=8, threads=1
- `GetMergeOperands()` latency increases 682 micros -> 800 micros (+17%)
The regression disappears into the noise (<1% difference) if we remove the `Reset()` loop and the size counting loop. The former is arguably needed regardless of this PR as the convention in `Get()` and `MultiGet()` is to `Reset()` the input `PinnableSlice`s at the start. The latter could be optimized to count the size as we accumulate operands rather than after the fact.
Best improvement is when a key has large operands and high concurrency:
- Parameters: num=4 (implying 4096 operands per key), value_sz=2KB, threads=32
- `GetMergeOperands()` latency decreases 11492 micros -> 437 micros (-96%).
Reviewed By: cbi42
Differential Revision: D38336578
Pulled By: ajkr
fbshipit-source-id: 48146d127e04cb7f2d4d2939a2b9dff3aba18258
Summary:
If a db_bench process gets hung or runaway on a machine, that
could prevent regression_test.sh from ever making progress. To fix that,
regression_test.sh will now kill any db_bench process that is >12 hours
old. Also made this more reliable by not using string matching (grep) to
get db_bench process IDs.
I also had to make some other updates to get local runs working
reliably:
* Fix some quoting hell and other dubious complexity with db_bench_cmd
* Only save a DB for re-use when building it passes
* Report failed command in more cases
* Add safeguards against "rm -rf ."
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10441
Test Plan:
manual (local and remote), with temporary changes e.g. to have
a manageable age threshold etc.
Reviewed By: riversand963
Differential Revision: D38285537
Pulled By: pdillinger
fbshipit-source-id: 4d598876aedc38ac4bd9d8ddf32c5995d8e44db8
Summary:
The secondary cache is randomly disabled or enabled with CompressedSecondaryCache.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10442
Test Plan: - To test that the CompressedSecondaryCache is used and the stress test runs successfully, run `make -j24 CRASH_TEST_EXT_ARGS=—duration=960 blackbox_crash_test `
Reviewed By: anand1976
Differential Revision: D38290796
Pulled By: gitbw95
fbshipit-source-id: bb7027b39e0ed9c0c62835abe09e759898130ec8
Summary:
deleterandom tests are too fast to get good signal, e.g.
--deletes=31250 in 0.170 seconds vs. --reads=1500000 in 288.491
seconds for readrandom. Removing the special handling (unknown
motivation in faa7eb3b99) should suffice.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10437
Test Plan: watch continuous results
Reviewed By: ltamasi
Differential Revision: D38261185
Pulled By: pdillinger
fbshipit-source-id: 0f1b1b19efccda5689027d36cc2f01307f36031d
Summary:
RocksDB's `Cache` abstraction currently supports two priority levels for items: high (used for frequently accessed/highly valuable SST metablocks like index/filter blocks) and low (used for SST data blocks). Blobs are typically lower-value targets for caching than data blocks, since 1) with BlobDB, data blocks containing blob references conceptually form an index structure which has to be consulted before we can read the blob value, and 2) cached blobs represent only a single key-value, while cached data blocks generally contain multiple KVs. Since we would like to make it possible to use the same backing cache for the block cache and the blob cache, it would make sense to add a new, lower-than-low cache priority level (bottom level) for blobs so data blocks are prioritized over them.
This task is a part of https://github.com/facebook/rocksdb/issues/10156
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10309
Reviewed By: ltamasi
Differential Revision: D38211655
Pulled By: gangliao
fbshipit-source-id: 65ef33337db4d85277cc6f9782d67c421ad71dd5
Summary:
In this PR we bring ClockCache closer to production quality. We implement the following changes:
1. Fixed a few bugs in ClockCache.
2. ClockCache now fully supports ``strict_capacity_limit == false``: When an insertion over capacity is commanded, we allocate a handle separately from the hash table.
3. ClockCache now runs on almost every test in cache_test. The only exceptions are a test where either the LRU policy is required, and a test that dynamically increases the table capacity.
4. ClockCache now supports dynamically decreasing capacity via SetCapacity. (This is easy: we shrink the capacity upper bound and run the clock algorithm.)
5. Old FastLRUCache tests in lru_cache_test.cc are now also used on ClockCache.
As a byproduct of 1. and 2. we are able to turn on ClockCache in the stress tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10418
Test Plan:
- ``make -j24 USE_CLANG=1 COMPILE_WITH_ASAN=1 COMPILE_WITH_UBSAN=1 check``
- ``make -j24 USE_CLANG=1 COMPILE_WITH_TSAN=1 check``
- ``make -j24 USE_CLANG=1 COMPILE_WITH_ASAN=1 COMPILE_WITH_UBSAN=1 CRASH_TEST_EXT_ARGS="--duration=960 --cache_type=clock_cache" blackbox_crash_test_with_atomic_flush``
- ``make -j24 USE_CLANG=1 COMPILE_WITH_TSAN=1 CRASH_TEST_EXT_ARGS="--duration=960 --cache_type=clock_cache" blackbox_crash_test_with_atomic_flush``
Reviewed By: pdillinger
Differential Revision: D38170673
Pulled By: guidotag
fbshipit-source-id: 508987b9dc9d9d68f1a03eefac769820b680340a
Summary:
If user runs `db_bench` with `-perf_level=2` or higher, db_bench should
print perf context after each of all benchmarks.
Or make `-perf_level` a per-benchmark switch.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10396
Test Plan: ./db_bench -benchmarks=fillseq,readseq -perf_level=2
Reviewed By: ajkr
Differential Revision: D38016324
Pulled By: riversand963
fbshipit-source-id: d83ea4abc34d40ffea394ca6abf0814bc5c0a2e0
Summary:
To help service owners to manage their memory budget effectively, we have been working towards counting all major memory users inside RocksDB towards a single global memory limit (see e.g. https://github.com/facebook/rocksdb/wiki/Write-Buffer-Manager#cost-memory-used-in-memtable-to-block-cache). The global limit is specified by the capacity of the block-based table's block cache, and is technically implemented by inserting dummy entries ("reservations") into the block cache. The goal of this task is to support charging the memory usage of the new blob cache against this global memory limit when the backing cache of the blob cache and the block cache are different.
This PR is a part of https://github.com/facebook/rocksdb/issues/10156
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10321
Reviewed By: ltamasi
Differential Revision: D37913590
Pulled By: gangliao
fbshipit-source-id: eaacf23907f82dc7d18964a3f24d7039a2937a72
Summary:
After branch 7.5.fb branch is cut, following release process, upgrade version number to 7.6 and add 7.5.fb to format compatibility check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10376
Test Plan: Watch CI
Reviewed By: ajkr
Differential Revision: D37927694
fbshipit-source-id: 71b37ae55ebb7c95a1bcc0d7eee643d6ba5f8461
Summary:
Many workloads have temporal locality, where recently written items are read back in a short period of time. When using remote file systems, this is inefficient since it involves network traffic and higher latencies. Because of this, we would like to support prepopulating the blob cache during flush.
This task is a part of https://github.com/facebook/rocksdb/issues/10156
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10298
Reviewed By: ltamasi
Differential Revision: D37908743
Pulled By: gangliao
fbshipit-source-id: 9feaed234bc719d38f0c02975c1ad19fa4bb37d1
Summary:
ClockCache is still in experimental stage, and currently fails some pre-release fbcode tests. See https://www.internalfb.com/diff/D37772011. API calls to construct ClockCache are done via the function NewClockCache. For now, NewClockCache calls will return an LRUCache (with appropriate arguments), which is stable.
The idea that NewClockCache returns nullptr was also floated, but this would be interpreted as unsupported cache, and a default LRUCache would be constructed instead, potentially causing a performance regression that is harder to identify.
A new version of the NewClockCache function was created for our internal tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10351
Test Plan: ``make -j24 check`` and re-run the pre-release tests.
Reviewed By: pdillinger
Differential Revision: D37802685
Pulled By: guidotag
fbshipit-source-id: 0a8d10612ff21e576f7360cb13e20bc36e244972
Summary:
as title.
Test plan
- make check
- CI on PR
- TEST_TMPDIR=/dev/shm make crash_test_with_multiops_wp_txn (tested with successful run)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10350
Reviewed By: ajkr
Differential Revision: D37792872
Pulled By: riversand963
fbshipit-source-id: ff064093b7f715d0acf387af2e3ae87b1278b52b
Summary:
Previously the version was displayed as $major.$minor
This changes it to $major.$minor.$path
This also adds the git hash for the time from which RocksDB was built to the end of report.tsv. I confirmed that benchmark_log_tool.py still parses it and that the people
who consume/graph these results are OK with it.
Example output:
ops_sec mb_sec lsm_sz blob_sz c_wgb w_amp c_mbps c_wsecs c_csecs b_rgb b_wgb usec_op p50 p99 p99.9 p99.99 pmax uptime stall% Nstall u_cpu s_cpu rss test date version job_id githash
609488 244.1 1GB 0.0GB, 1.4 0.7 93.3 39 38 0 0 1.6 1.0 4 15 26 5365 15 0.0 0 0.1 0.0 0.5 fillseq.wal_disabled.v400 2022-06-29T13:36:05 7.5.0 6115254416
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10277
Test Plan: Run it
Reviewed By: jay-zhuang
Differential Revision: D37532418
Pulled By: mdcallag
fbshipit-source-id: 55e472640d51265819b228d3373c9fa9b62b660d
Summary:
This adds --undefok to support use of this script with BlobDB for db_bench versions prior
to 7.5 when the options land in a release.
While there is a limit to how far back this script can go WRT backwards compatiblity,
this is an easy change to support early 7.x releases.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10276
Test Plan: Run it with versions of db_bench that do not and then do support these options
Reviewed By: gangliao
Differential Revision: D37529299
Pulled By: mdcallag
fbshipit-source-id: 7bb1feec5c68760e6d64792c585bfbde4f5e52d8
Summary:
This is the initial step in the development of a lock-free clock cache. This PR includes the base hash table design (which we mostly ported over from FastLRUCache) and the clock eviction algorithm. Importantly, it's still _not_ lock-free---all operations use a shard lock. Besides the locking, there are other features left as future work:
- Remove keys from the handles. Instead, use 128-bit bijective hashes of them for handle comparisons, probing (we need two 32-bit hashes of the key for double hashing) and sharding (we need one 6-bit hash).
- Remove the clock_usage_ field, which is updated on every lookup. Even if it were atomically updated, it could cause memory invalidations across cores.
- Middle insertions into the clock list.
- A test that exercises the clock eviction policy.
- Update the Java API of ClockCache and Java calls to C++.
Along the way, we improved the code and comments quality of FastLRUCache. These changes are relatively minor.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10273
Test Plan: ``make -j24 check``
Reviewed By: pdillinger
Differential Revision: D37522461
Pulled By: guidotag
fbshipit-source-id: 3d70b737dbb70dcf662f00cef8c609750f083943
Summary:
Before this PR, when user-defined timestamp is enabled, db_stress disables compaction filter.
This is no longer necessary after this PR, since the `DbStressCompactionFilter` is now aware of
the presence of timestamps.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10259
Test Plan: TEST_TMPDIR=/dev/shm make crash_test_with_ts
Reviewed By: ajkr
Differential Revision: D37459692
Pulled By: riversand963
fbshipit-source-id: 8fe62e90a63bd9317fe1bb95a2b4984080c9e5ef
Summary:
Need to disable it for now as CI is failing, particularly `MultiOpsTxnsStressTest`. Investigation details in internal task T124324915. This PR disables mempurge more widely than `MultiOpsTxnsStressTest` until we know the issue is contained to that particular test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10252
Reviewed By: riversand963
Differential Revision: D37432948
Pulled By: ajkr
fbshipit-source-id: d0cf5b0e0ec7c3142c382a0347f35a4c34f4607a
Summary:
In order to facilitate correctness and performance testing, we would like to add the new blob cache to our stress test tool `db_stress` and our continuously running crash test script `db_crashtest.py`, as well as our synthetic benchmarking tool `db_bench` and the BlobDB performance testing script `run_blob_bench.sh`.
As part of this task, we would also like to utilize these benchmarking tools to get some initial performance numbers about the effectiveness of caching blobs.
This PR is a part of https://github.com/facebook/rocksdb/issues/10156
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10202
Reviewed By: ltamasi
Differential Revision: D37325739
Pulled By: gangliao
fbshipit-source-id: deb65d0d414502270dd4c324d987fd5469869fa8
Summary:
There was a bug in the MultiGet enhancement in https://github.com/facebook/rocksdb/issues/9899 with data
block hash index, which was not caught because data block hash index was
never added to stress tests. This change fixes both issues.
Fixes https://github.com/facebook/rocksdb/issues/10186
I intend to pick this into the 7.4.0 release candidate
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10220
Test Plan:
Failure quickly reproduces in crash test with
kDataBlockBinaryAndHash, and does not seem to with the fix. Reproducing
the failure with a unit test I believe would be too tricky and fragile
to be worthwhile.
Reviewed By: anand1976
Differential Revision: D37315647
Pulled By: pdillinger
fbshipit-source-id: 9f648265bba867275edc752f7a56611a59401cba
Summary:
In FastLRUCache, we replace the current chained per-shard hash table by an open-addressing hash table. In particular, this allows us to preallocate all handles.
Because all handles are preallocated, this implementation doesn't support strict_capacity_limit = false (i.e., allowing insertions beyond the predefined capacity). This clashes with current assumptions of some tests, namely two tests in cache_test and the crash tests. We have disabled these for now.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10194
Test Plan: ``make -j24 check``
Reviewed By: pdillinger
Differential Revision: D37296770
Pulled By: guidotag
fbshipit-source-id: 232ff1b8260331d868ebf4e3e5d8ad709390b0ad
Summary:
There is currently no caching mechanism for blobs, which is not ideal especially when the database resides on remote storage (where we cannot rely on the OS page cache). As part of this task, we would like to make it possible for the application to configure a blob cache.
In this task, we formally introduced the blob source to RocksDB. BlobSource is a new abstraction layer that provides universal access to blobs, regardless of whether they are in the blob cache, secondary cache, or (remote) storage. Depending on user settings, it always fetch blobs from multi-tier cache and storage with minimal cost.
Note: The new `MultiGetBlob()` implementation is not included in the current PR. To go faster, we aim to create a separate PR for it in parallel!
This PR is a part of https://github.com/facebook/rocksdb/issues/10156
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10198
Reviewed By: ltamasi
Differential Revision: D37294735
Pulled By: gangliao
fbshipit-source-id: 9cb50422d9dd1bc03798501c2778b6c7520c7a1e
Summary:
**Context/Summary:**
https://github.com/facebook/rocksdb/pull/9424 added rate-limiting support for user reads, which does not include batched `MultiGet()`s that call `RandomAccessFileReader::MultiRead()`. The reason is that it's harder (compared with RandomAccessFileReader::Read()) to implement the ideal rate-limiting where we first call `RateLimiter::RequestToken()` for allowed bytes to multi-read and then consume those bytes by satisfying as many requests in `MultiRead()` as possible. For example, it can be tricky to decide whether we want partially fulfilled requests within one `MultiRead()` or not.
However, due to a recent urgent user request, we decide to pursue an elementary (but a conditionally ineffective) solution where we accumulate enough rate limiter requests toward the total bytes needed by one `MultiRead()` before doing that `MultiRead()`. This is not ideal when the total bytes are huge as we will actually consume a huge bandwidth from rate-limiter causing a burst on disk. This is not what we ultimately want with rate limiter. Therefore a follow-up work is noted through TODO comments.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10159
Test Plan:
- Modified existing unit test `DBRateLimiterOnReadTest/DBRateLimiterOnReadTest.NewMultiGet`
- Traced the underlying system calls `io_uring_enter` and verified they are 10 seconds apart from each other correctly under the setting of `strace -ftt -e trace=io_uring_enter ./db_bench -benchmarks=multireadrandom -db=/dev/shm/testdb2 -readonly -num=50 -threads=1 -multiread_batched=1 -batch_size=100 -duration=10 -rate_limiter_bytes_per_sec=200 -rate_limiter_refill_period_us=1000000 -rate_limit_bg_reads=1 -disable_auto_compactions=1 -rate_limit_user_ops=1` where each `MultiRead()` read about 2000 bytes (inspected by debugger) and the rate limiter grants 200 bytes per seconds.
- Stress test:
- Verified `./db_stress (-test_cf_consistency=1/test_batches_snapshots=1) -use_multiget=1 -cache_size=1048576 -rate_limiter_bytes_per_sec=10241024 -rate_limit_bg_reads=1 -rate_limit_user_ops=1` work
Reviewed By: ajkr, anand1976
Differential Revision: D37135172
Pulled By: hx235
fbshipit-source-id: 73b8e8f14761e5d4b77235dfe5d41f4eea968bcd
Summary:
Added an option, `WriteOptions::protection_bytes_per_key`, that controls how many bytes per key we use for integrity protection in `WriteBatch`. It takes effect when `WriteBatch::GetProtectionBytesPerKey() == 0`.
Currently the only supported value is eight. Invoking a user API with it set to any other nonzero value will result in `Status::NotSupported` returned to the user.
There is also a bug fix for integrity protection with `inplace_callback`, where we forgot to take into account the possible change in varint length when calculating KV checksum for the final encoded buffer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10037
Test Plan:
- Manual
- Set default value of `WriteOptions::protection_bytes_per_key` to eight and ran `make check -j24`
- Enabled in MyShadow for 1+ week
- Automated
- Unit tests have a `WriteMode` that enables the integrity protection via `WriteOptions`
- Crash test - in most cases, use `WriteOptions::protection_bytes_per_key` to enable integrity protection
Reviewed By: cbi42
Differential Revision: D36614569
Pulled By: ajkr
fbshipit-source-id: 8650087ceac9b61b560f1e5fafe5e1baf9c725fb
Summary:
In https://github.com/facebook/rocksdb/issues/9535, release 7.0, we hid the old block-based filter from being created using
the public API, because of its inefficiency. Although we normally maintain read compatibility
on old DBs forever, filters are not required for reading a DB, only for optimizing read
performance. Thus, it should be acceptable to remove this code and the substantial
maintenance burden it carries as useful features are developed and validated (such
as user timestamp).
This change completely removes the code for reading and writing the old block-based
filters, net removing about 1370 lines of code no longer needed. Options removed from
testing / benchmarking tools. The prior existence is only evident in a couple of places:
* `CacheEntryRole::kDeprecatedFilterBlock` - We can update this public API enum in
a major release to minimize source code incompatibilities.
* A warning is logged when an old table file is opened that used the old block-based
filter. This is provided as a courtesy, and would be a pain to unit test, so manual testing
should suffice. Unfortunately, sst_dump does not tell you whether a file uses
block-based filter, and the structure of the code makes it very difficult to fix.
* To detect that case, `kObsoleteFilterBlockPrefix` (renamed from `kFilterBlockPrefix`)
for metaindex is maintained (for now).
Other notes:
* In some cases where numbers are associated with filter configurations, we have had to
update the assigned numbers so that they all correspond to something that exists.
* Fixed potential stat counting bug by assuming `filter_checked = false` for cases
like `filter == nullptr` rather than assuming `filter_checked = true`
* Removed obsolete `block_offset` and `prefix_extractor` parameters from several
functions.
* Removed some unnecessary checks `if (!table_prefix_extractor() && !prefix_extractor)`
because the caller guarantees the prefix extractor exists and is compatible
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10184
Test Plan:
tests updated, manually test new warning in LOG using base version to
generate a DB
Reviewed By: riversand963
Differential Revision: D37212647
Pulled By: pdillinger
fbshipit-source-id: 06ee020d8de3b81260ffc36ad0c1202cbf463a80
Summary:
Fix existing usage of non-ASCII and add a check to prevent
future use. Added `-n` option to greps to provide line numbers.
Alternative to https://github.com/facebook/rocksdb/issues/10147
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10164
Test Plan:
used new checker to find & fix cases, manually check
db_bench output is preserved
Reviewed By: akankshamahajan15
Differential Revision: D37148792
Pulled By: pdillinger
fbshipit-source-id: 68c8b57e7ab829369540d532590bf756938855c7
