Tag:
Branch:
Tree:
ff1cc8a63e
main
oxigraph-8.1.1
oxigraph-8.3.2
oxigraph-main
${ noResults }
1470 Commits (ff1cc8a63e64684bcdfab099e3728da51d7fb2e3)
Author | SHA1 | Message | Date |
---|---|---|---|
Gang Liao | 0b6bc101ba |
Charge blob cache usage against the global memory limit (#10321)
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 |
2 years ago |
Jay Zhuang | a3acf2ef87 |
Add seqno to time mapping (#10338)
Summary: Which will be used for tiered storage to preclude hot data from compacting to the cold tier (the last level). Internally, adding seqno to time mapping. A periodic_task is scheduled to record the current_seqno -> current_time in certain cadence. When memtable flush, the mapping informaiton is stored in sstable property. During compaction, the mapping information are merged and get the approximate time of sequence number, which is used to determine if a key is recently inserted or not and preclude it from the last level if it's recently inserted (within the `preclude_last_level_data_seconds`). Pull Request resolved: https://github.com/facebook/rocksdb/pull/10338 Test Plan: CI Reviewed By: siying Differential Revision: D37810187 Pulled By: jay-zhuang fbshipit-source-id: 6953be7a18a99de8b1cb3b162d712f79c2b4899f |
2 years ago |
sdong | c8b20d469d |
Make InternalKeyComparator not configurable (#10342)
Summary: InternalKeyComparator is an internal class which is a simple wrapper of Comparator. https://github.com/facebook/rocksdb/pull/8336 made Comparator customizeable. As a side effect, internal key comparator was made configurable too. This introduces overhead to this simple wrapper. For example, every InternalKeyComparator will have an std::vector attached to it, which consumes memory and possible allocation overhead too. We remove InternalKeyComparator from being customizable by making InternalKeyComparator not a subclass of Comparator. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10342 Test Plan: Run existing CI tests and make sure it doesn't fail Reviewed By: riversand963 Differential Revision: D37771351 fbshipit-source-id: 917256ee04b2796ed82974549c734fb6c4d8ccee |
2 years ago |
sdong | 769b156e65 |
Remove customized naming from InternalKeyComparator (#10343)
Summary: InternalKeyComparator is a thin wrapper around user comparator. Storing a string for name is relatively expensive to this small wrapper for both CPU and memory usage. Try to remove it. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10343 Test Plan: Run existing tests Reviewed By: ajkr Differential Revision: D37772469 fbshipit-source-id: d2d106a8d022193058fd7f6b220108e3d94aca34 |
2 years ago |
Peter Dillinger | e6c5e0ab9a |
Have Cache use Status::MemoryLimit (#10262)
Summary:
I noticed it would clean up some things to have Cache::Insert()
return our MemoryLimit Status instead of Incomplete for the case in
which the capacity limit is reached. I suspect this fixes some existing but
unknown bugs where this Incomplete could be confused with other uses
of Incomplete, especially no_io cases. This is the most suspicious case I
noticed, but was not able to reproduce a bug, in part because the existing
code is not covered by unit tests (FIXME added):
|
3 years ago |
Akanksha Mahajan | 2acbf386a3 |
Provide support for direct_reads with async_io (#10197)
Summary: Provide support for use_direct_reads with async_io. TestPlan: - Updated unit tests - db_bench: Results in https://github.com/facebook/rocksdb/pull/10197#issuecomment-1159239420 - db_stress ``` export CRASH_TEST_EXT_ARGS=" --async_io=1 --use_direct_reads=1" make crash_test -j ``` - Ran db_bench on previous RocksDB version before any async_io implementation (as there have many changes in different PRs in this area) https://github.com/facebook/rocksdb/pull/10197#issuecomment-1160781563. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10197 Reviewed By: anand1976 Differential Revision: D37255646 Pulled By: akankshamahajan15 fbshipit-source-id: fec61ae15bf4d625f79dea56e4f86e0e307ba920 |
3 years ago |
Changyu Bi | f9cfc6a808 |
Updated NewDataBlockIterator to not fetch compression dict for non-da… (#10310)
Summary: …ta blocks During MyShadow testing, ajkr helped me find out that with partitioned index and dictionary compression enabled, `PartitionedIndexIterator::InitPartitionedIndexBlock()` spent considerable amount of time (1-2% CPU) on fetching uncompression dictionary. Fetching uncompression dict was not needed since the index blocks were not compressed (and even if they were, they use empty dictionary). This should only affect use cases with partitioned index, dictionary compression and without uncompression dictionary pinned. This PR updates NewDataBlockIterator to not fetch uncompression dictionary when it is not for data blocks. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10310 Test Plan: 1. `make check` 2. Perf benchmark: 1.5% (143950 -> 146176) improvement in op/sec for partitioned index + dict compression benchmark. For default config without partitioned index and without dict compression, there is no regression in readrandom perf from multiple runs of db_bench. ``` # Set up for partitioned index with dictionary compression TEST_TMPDIR=/dev/shm ./db_bench_main -benchmarks=filluniquerandom,compact -max_background_jobs=24 -memtablerep=vector -allow_concurrent_memtable_write=false -partition_index=true -compression_max_dict_bytes=16384 -compression_zstd_max_train_bytes=1638400 # Pre PR TEST_TMPDIR=/dev/shm ./db_bench_main -use_existing_db=true -benchmarks=readrandom[-X50] -partition_index=true readrandom [AVG 50 runs] : 143950 (± 1108) ops/sec; 15.9 (± 0.1) MB/sec readrandom [MEDIAN 50 runs] : 144406 ops/sec; 16.0 MB/sec # Post PR TEST_TMPDIR=/dev/shm ./db_bench_opt -use_existing_db=true -benchmarks=readrandom[-X50] -partition_index=true readrandom [AVG 50 runs] : 146176 (± 1121) ops/sec; 16.2 (± 0.1) MB/sec readrandom [MEDIAN 50 runs] : 146014 ops/sec; 16.2 MB/sec # Set up for no partitioned index and no dictionary compression TEST_TMPDIR=/dev/shm/baseline ./db_bench_main -benchmarks=filluniquerandom,compact -max_background_jobs=24 -memtablerep=vector -allow_concurrent_memtable_write=false # Pre PR TEST_TMPDIR=/dev/shm/baseline/ ./db_bench_main --use_existing_db=true "--benchmarks=readrandom[-X50]" readrandom [AVG 50 runs] : 158546 (± 1000) ops/sec; 17.5 (± 0.1) MB/sec readrandom [MEDIAN 50 runs] : 158280 ops/sec; 17.5 MB/sec # Post PR TEST_TMPDIR=/dev/shm/baseline/ ./db_bench_opt --use_existing_db=true "--benchmarks=readrandom[-X50]" readrandom [AVG 50 runs] : 161061 (± 1520) ops/sec; 17.8 (± 0.2) MB/sec readrandom [MEDIAN 50 runs] : 161596 ops/sec; 17.9 MB/sec ``` Reviewed By: ajkr Differential Revision: D37631358 Pulled By: cbi42 fbshipit-source-id: 6ca2665e270e63871968e061ba4a99d3136785d9 |
3 years ago |
Levi Tamasi | c73d2a9d18 |
Add API for writing wide-column entities (#10242)
Summary: The patch builds on https://github.com/facebook/rocksdb/pull/9915 and adds a new API called `PutEntity` that can be used to write a wide-column entity to the database. The new API is added to both `DB` and `WriteBatch`. Note that currently there is no way to retrieve these entities; more precisely, all read APIs (`Get`, `MultiGet`, and iterator) return `NotSupported` when they encounter a wide-column entity that is required to answer a query. Read-side support (as well as other missing functionality like `Merge`, compaction filter, and timestamp support) will be added in later PRs. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10242 Test Plan: `make check` Reviewed By: riversand963 Differential Revision: D37369748 Pulled By: ltamasi fbshipit-source-id: 7f5e412359ed7a400fd80b897dae5599dbcd685d |
3 years ago |
Bo Wang | 8e63d90ff8 |
Pass rate_limiter_priority through filter block reader functions to FS (#10251)
Summary: With https://github.com/facebook/rocksdb/pull/9996 , we can pass the rate_limiter_priority to FS for most cases. This PR is to update the code path for filter block reader. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10251 Test Plan: Current unit tests should pass. Reviewed By: pdillinger Differential Revision: D37427667 Pulled By: gitbw95 fbshipit-source-id: 1ce5b759b136efe4cfa48a6b97e2f837ff087433 |
3 years ago |
Bo Wang | c073ed7601 |
Fix typo in comments and code (#10233)
Summary: Fix typo in comments and code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10233 Test Plan: Existing unit tests should pass. Reviewed By: jay-zhuang, anand1976 Differential Revision: D37356702 Pulled By: gitbw95 fbshipit-source-id: 32c019adcc6dcc95a9882b38147a310091368e51 |
3 years ago |
Peter Dillinger | 84210c9489 |
Add data block hash index to crash test, fix MultiGet issue (#10220)
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 |
3 years ago |
Peter Dillinger | fff302d989 |
More testing w/prefix extractor, small refactor (#10122)
Summary: There was an interesting code path not covered by testing that is difficult to replicate in a unit test, which is now covered using a sync point. Specifically, the case of table_prefix_extractor == null and !need_upper_bound_check in `BlockBasedTable::PrefixMayMatch`, which can happen if table reader is open before extractor is registered with global object registry, but is later registered and re-set with SetOptions. (We don't have sufficient testing control over object registry to set that up repeatedly.) Also, this function has been renamed to `PrefixRangeMayMatch` for clarity vs. other functions that are not the same. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10122 Test Plan: unit tests expanded Reviewed By: siying Differential Revision: D36944834 Pulled By: pdillinger fbshipit-source-id: 9e52d9da1929a3e42bbc230fcdc3599949de7bdb |
3 years ago |
Peter Dillinger | 126c223714 |
Remove deprecated block-based filter (#10184)
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 |
3 years ago |
anand76 | a6691d0f65 |
Update stats to help users estimate MultiGet async IO impact (#10182)
Summary: Add a couple of stats to help users estimate the impact of potential MultiGet perf improvements - 1. NUM_LEVEL_READ_PER_MULTIGET - A histogram stat for number of levels that required MultiGet to read from a file 2. MULTIGET_COROUTINE_COUNT - A ticker stat to count the number of times the coroutine version of MultiGetFromSST was used The NUM_DATA_BLOCKS_READ_PER_LEVEL stat is obsoleted as it doesn't provide useful information for MultiGet optimization. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10182 Reviewed By: akankshamahajan15 Differential Revision: D37213296 Pulled By: anand1976 fbshipit-source-id: 5d2b7708017c0e278578ae4bffac3926f6530efb |
3 years ago |
Akanksha Mahajan | 8353ae8b27 |
Add few optimizations in async_io for short scans (#10140)
Summary: This PR adds few optimizations for async_io for shorter scans. 1. If async_io is enabled, seek would create FilePrefetchBuffer object to fetch the data asynchronously. However `FilePrefetchbuffer::num_file_reads_` wasn't taken into consideration if it calls Next after Seek and would go for Prefetching. This PR fixes that and Next will go for prefetching only if `FilePrefetchbuffer::num_file_reads_` is greater than 2 along with if blocks are sequential. This scenario is only for implicit auto readahead. 2. For seek, when it calls TryReadFromCacheAsync to poll it makes async call as well because TryReadFromCacheAsync flow wasn't changed. So I updated to return after poll instead of further prefetching any data. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10140 Test Plan: 1. Added a unit test 2. Ran crash_test with async_io = 1 to make sure nothing crashes. Reviewed By: anand1976 Differential Revision: D37042242 Pulled By: akankshamahajan15 fbshipit-source-id: b8e6b7cb2ee0886f37a8f53951948b9084e8ffda |
3 years ago |
Gang Liao | cba398df8a |
Add blob cache option in the column family options (#10155)
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. This PR is a part of https://github.com/facebook/rocksdb/issues/10156 Pull Request resolved: https://github.com/facebook/rocksdb/pull/10155 Reviewed By: ltamasi Differential Revision: D37150819 Pulled By: gangliao fbshipit-source-id: b807c7916ea5d411588128f8e22a49f171388fe2 |
3 years ago |
Hui Xiao | d665afdbf3 |
Account memory of FileMetaData in global memory limit (#9924)
Summary: **Context/Summary:** As revealed by heap profiling, allocation of `FileMetaData` for [newly created file added to a Version](https://github.com/facebook/rocksdb/pull/9924/files#diff-a6aa385940793f95a2c5b39cc670bd440c4547fa54fd44622f756382d5e47e43R774) can consume significant heap memory. This PR is to account that toward our global memory limit based on block cache capacity. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9924 Test Plan: - Previous `make check` verified there are only 2 places where the memory of the allocated `FileMetaData` can be released - New unit test `TEST_P(ChargeFileMetadataTestWithParam, Basic)` - db bench (CPU cost of `charge_file_metadata` in write and compact) - **write micros/op: -0.24%** : `TEST_TMPDIR=/dev/shm/testdb ./db_bench -benchmarks=fillseq -db=$TEST_TMPDIR -charge_file_metadata=1 (remove this option for pre-PR) -disable_auto_compactions=1 -write_buffer_size=100000 -num=4000000 | egrep 'fillseq'` - **compact micros/op -0.87%** : `TEST_TMPDIR=/dev/shm/testdb ./db_bench -benchmarks=fillseq -db=$TEST_TMPDIR -charge_file_metadata=1 -disable_auto_compactions=1 -write_buffer_size=100000 -num=4000000 -numdistinct=1000 && ./db_bench -benchmarks=compact -db=$TEST_TMPDIR -use_existing_db=1 -charge_file_metadata=1 -disable_auto_compactions=1 | egrep 'compact'` table 1 - write #-run | (pre-PR) avg micros/op | std micros/op | (post-PR) micros/op | std micros/op | change (%) -- | -- | -- | -- | -- | -- 10 | 3.9711 | 0.264408 | 3.9914 | 0.254563 | 0.5111933721 20 | 3.83905 | 0.0664488 | 3.8251 | 0.0695456 | -0.3633711465 40 | 3.86625 | 0.136669 | 3.8867 | 0.143765 | 0.5289363078 80 | 3.87828 | 0.119007 | 3.86791 | 0.115674 | **-0.2673865734** 160 | 3.87677 | 0.162231 | 3.86739 | 0.16663 | **-0.2419539978** table 2 - compact #-run | (pre-PR) avg micros/op | std micros/op | (post-PR) micros/op | std micros/op | change (%) -- | -- | -- | -- | -- | -- 10 | 2,399,650.00 | 96,375.80 | 2,359,537.00 | 53,243.60 | -1.67 20 | 2,410,480.00 | 89,988.00 | 2,433,580.00 | 91,121.20 | 0.96 40 | 2.41E+06 | 121811 | 2.39E+06 | 131525 | **-0.96** 80 | 2.40E+06 | 134503 | 2.39E+06 | 108799 | **-0.78** - stress test: `python3 tools/db_crashtest.py blackbox --charge_file_metadata=1 --cache_size=1` killed as normal Reviewed By: ajkr Differential Revision: D36055583 Pulled By: hx235 fbshipit-source-id: b60eab94707103cb1322cf815f05810ef0232625 |
3 years ago |
Peter Dillinger | d3a3b02134 |
Fix bug with kHashSearch and changing prefix_extractor with SetOptions (#10128)
Summary: When opening an SST file created using index_type=kHashSearch, the *current* prefix_extractor would be saved, and used with hash index if the *new current* prefix_extractor at query time is compatible with the SST file. This is a problem if the prefix_extractor at SST open time is not compatible but SetOptions later changes (back) to one that is compatible. This change fixes that by using the known compatible (or missing) prefix extractor we save for use with prefix filtering. Detail: I have moved the InternalKeySliceTransform wrapper to avoid some indirection and remove unnecessary fields. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10128 Test Plan: expanded unit test (using some logic from https://github.com/facebook/rocksdb/issues/10122) that fails before fix and probably covers some other previously uncovered cases. Reviewed By: siying Differential Revision: D36955738 Pulled By: pdillinger fbshipit-source-id: 0c78a6b0d24054ef2f3cb237bf010c1c5589fb10 |
3 years ago |
Peter Dillinger | 4f78f9699b |
Refactor: Add BlockTypes to make them imply C++ type in block cache (#10098)
Summary: We have three related concepts: * BlockType: an internal enum conceptually indicating a type of SST file block * CacheEntryRole: a user-facing enum for categorizing block cache entries, which is also involved in associated cache entries with an appropriate deleter. Can include categories for non-block cache entries (e.g. memory reservations). * TBlocklike: a C++ type for the actual type behind a void* cache entry. We had some existing code ugliness because BlockType did not imply TBlocklike, because of various kinds of "filter" block. This refactoring fixes that with new BlockTypes. More clean-up can come in later work. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10098 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D36897945 Pulled By: pdillinger fbshipit-source-id: 3ae496b5caa81e0a0ed85e873eb5b525e2d9a295 |
3 years ago |
Yanqin Jin | 3e02c6e05a |
Point-lookup returns timestamps of Delete and SingleDelete (#10056)
Summary: If caller specifies a non-null `timestamp` argument in `DB::Get()` or a non-null `timestamps` in `DB::MultiGet()`, RocksDB will return the timestamps of the point tombstones. Note: DeleteRange is still unsupported. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10056 Test Plan: make check Reviewed By: ltamasi Differential Revision: D36677956 Pulled By: riversand963 fbshipit-source-id: 2d7af02cc7237b1829cd269086ea895a49d501ae |
3 years ago |
Hui Xiao | 4bdcc80192 |
Increase ChargeTableReaderTest/ChargeTableReaderTest.Basic error tolerance rate from 1% to 5% (#10113)
Summary: **Context:** https://github.com/facebook/rocksdb/pull/9748 added support to charge table reader memory to block cache. In the test `ChargeTableReaderTest/ChargeTableReaderTest.Basic`, it estimated the table reader memory, calculated the expected number of table reader opened based on this estimation and asserted this number with actual number. The expected number of table reader opened calculated based on estimated table reader memory will not be 100% accurate and should have tolerance for error. It was previously set to 1% and recently encountered an assertion failure that `(opened_table_reader_num) <= (max_table_reader_num_capped_upper_bound), actual: 375 or 376 vs 374` where `opened_table_reader_num` is the actual opened one and `max_table_reader_num_capped_upper_bound` is the estimated opened one (=371 * 1.01). I believe it's safe to increase error tolerance from 1% to 5% hence there is this PR. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10113 Test Plan: - CI again succeeds. Reviewed By: ajkr Differential Revision: D36911556 Pulled By: hx235 fbshipit-source-id: 259687dd77b450fea0f5658a5b567a1d31d4b1f7 |
3 years ago |
tagliavini | 6c50082654 |
Remove code that only compiles for Visual Studio versions older than 2015 (#10065)
Summary: There are currently some preprocessor checks that assume support for Visual Studio versions older than 2015 (i.e., 0 < _MSC_VER < 1900), although we don't support them any more. We removed all code that only compiles on those older versions, except third-party/ files. The ROCKSDB_NOEXCEPT symbol is now obsolete, since it now always gets replaced by noexcept. We removed it. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10065 Reviewed By: pdillinger Differential Revision: D36721901 Pulled By: guidotag fbshipit-source-id: a2892d365ef53cce44a0a7d90dd6b72ee9b5e5f2 |
3 years ago |
Jay Zhuang | a96a4a2f7b |
Fix ApproximateOffsetOfCompressed test (#10048)
Summary: https://github.com/facebook/rocksdb/issues/9857 introduced new an option `use_zstd_dict_trainer`, which is stored in SST as text, e.g.: ``` ... zstd_max_train_bytes=0; enabled=0;... ``` it increased the sst size a little bit and cause `ApproximateOffsetOfCompressed` test to fail: ``` Value 7053 is not in range [4000, 7050] table/table_test.cc:4019: Failure Value of: Between(c.ApproximateOffsetOf("xyz"), 4000, 7050) ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/10048 Test Plan: verified the test pass after the change Reviewed By: cbi42 Differential Revision: D36643688 Pulled By: jay-zhuang fbshipit-source-id: bf12d211f6ae71937259ef21b1226bd06e8da717 |
3 years ago |
Akanksha Mahajan | a479c2c2b2 |
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032)
Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf |
3 years ago |
Akanksha Mahajan | 2db6a4a1d6 |
Seek parallelization (#9994)
Summary: The RocksDB iterator is a hierarchy of iterators. MergingIterator maintains a heap of LevelIterators, one for each L0 file and for each non-zero level. The Seek() operation naturally lends itself to parallelization, as it involves positioning every LevelIterator on the correct data block in the correct SST file. It lookups a level for a target key, to find the first key that's >= the target key. This typically involves reading one data block that is likely to contain the target key, and scan forward to find the first valid key. The forward scan may read more data blocks. In order to find the right data block, the iterator may read some metadata blocks (required for opening a file and searching the index). This flow can be parallelized. Design: Seek will be called two times under async_io option. First seek will send asynchronous request to prefetch the data blocks at each level and second seek will follow the normal flow and in FilePrefetchBuffer::TryReadFromCacheAsync it will wait for the Poll() to get the results and add the iterator to min_heap. - Status::TryAgain is passed down from FilePrefetchBuffer::PrefetchAsync to block_iter_.Status indicating asynchronous request has been submitted. - If for some reason asynchronous request returns error in submitting the request, it will fallback to sequential reading of blocks in one pass. - If the data already exists in prefetch_buffer, it will return the data without prefetching further and it will be treated as single pass of seek. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9994 Test Plan: - **Run Regressions.** ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 -use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` i) Previous release 7.0 run for normal prefetching with async_io disabled: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Set seed to 1652922591315307 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:09:51 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483080.466 micros/op 2 ops/sec 120.287 seconds 249 operations; 340.8 MB/s (249 of 249 found) ``` iii) db_bench with async_io enabled completed succesfully ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 -async_io=1 -adaptive_readahead=1 Set seed to 1652924062021732 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:34:22 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 553913.576 micros/op 1 ops/sec 120.199 seconds 217 operations; 293.6 MB/s (217 of 217 found) ``` - db_stress with async_io disabled completed succesfully ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` I**n Progress**: db_stress with async_io is failing and working on debugging/fixing it. Reviewed By: anand1976 Differential Revision: D36459323 Pulled By: akankshamahajan15 fbshipit-source-id: abb1cd944abe712bae3986ae5b16704b3338917c |
3 years ago |
anand76 | e015206dd6 |
Fix crash due to MultiGet async IO and direct IO (#10024)
Summary: MultiGet with async IO is not officially supported with Posix yet. Avoid a crash by using synchronous MultiRead when direct IO is enabled. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10024 Test Plan: Run db_crashtest.py manually Reviewed By: akankshamahajan15 Differential Revision: D36551053 Pulled By: anand1976 fbshipit-source-id: 72190418fa92dd0397e87825df618b12c9bdecda |
3 years ago |
Changyu Bi | cc23b46da1 |
Support using ZDICT_finalizeDictionary to generate zstd dictionary (#9857)
Summary:
An untrained dictionary is currently simply the concatenation of several samples. The ZSTD API, ZDICT_finalizeDictionary(), can improve such a dictionary's effectiveness at low cost. This PR changes how dictionary is created by calling the ZSTD ZDICT_finalizeDictionary() API instead of creating raw content dictionary (when max_dict_buffer_bytes > 0), and pass in all buffered uncompressed data blocks as samples.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9857
Test Plan:
#### db_bench test for cpu/memory of compression+decompression and space saving on synthetic data:
Set up: change the parameter [here](
|
3 years ago |
anand76 | 57997ddaaf |
Multi file concurrency in MultiGet using coroutines and async IO (#9968)
Summary: This PR implements a coroutine version of batched MultiGet in order to concurrently read from multiple SST files in a level using async IO, thus reducing the latency of the MultiGet. The API from the user perspective is still synchronous and single threaded, with the RocksDB part of the processing happening in the context of the caller's thread. In Version::MultiGet, the decision is made whether to call synchronous or coroutine code. A good way to review this PR is to review the first 4 commits in order - de773b3, 70c2f70, 10b50e1, and 377a597 - before reviewing the rest. TODO: 1. Figure out how to build it in CircleCI (requires some dependencies to be installed) 2. Do some stress testing with coroutines enabled No regression in synchronous MultiGet between this branch and main - ``` ./db_bench -use_existing_db=true --db=/data/mysql/rocksdb/prefix_scan -benchmarks="readseq,multireadrandom" -key_size=32 -value_size=512 -num=5000000 -batch_size=64 -multiread_batched=true -use_direct_reads=false -duration=60 -ops_between_duration_checks=1 -readonly=true -adaptive_readahead=true -threads=16 -cache_size=10485760000 -async_io=false -multiread_stride=40000 -statistics ``` Branch - ```multireadrandom : 4.025 micros/op 3975111 ops/sec 60.001 seconds 238509056 operations; 2062.3 MB/s (14767808 of 14767808 found)``` Main - ```multireadrandom : 3.987 micros/op 4013216 ops/sec 60.001 seconds 240795392 operations; 2082.1 MB/s (15231040 of 15231040 found)``` More benchmarks in various scenarios are given below. The measurements were taken with ```async_io=false``` (no coroutines) and ```async_io=true``` (use coroutines). For an IO bound workload (with every key requiring an IO), the coroutines version shows a clear benefit, being ~2.6X faster. For CPU bound workloads, the coroutines version has ~6-15% higher CPU utilization, depending on how many keys overlap an SST file. 1. Single thread IO bound workload on remote storage with sparse MultiGet batch keys (~1 key overlap/file) - No coroutines - ```multireadrandom : 831.774 micros/op 1202 ops/sec 60.001 seconds 72136 operations; 0.6 MB/s (72136 of 72136 found)``` Using coroutines - ```multireadrandom : 318.742 micros/op 3137 ops/sec 60.003 seconds 188248 operations; 1.6 MB/s (188248 of 188248 found)``` 2. Single thread CPU bound workload (all data cached) with ~1 key overlap/file - No coroutines - ```multireadrandom : 4.127 micros/op 242322 ops/sec 60.000 seconds 14539384 operations; 125.7 MB/s (14539384 of 14539384 found)``` Using coroutines - ```multireadrandom : 4.741 micros/op 210935 ops/sec 60.000 seconds 12656176 operations; 109.4 MB/s (12656176 of 12656176 found)``` 3. Single thread CPU bound workload with ~2 key overlap/file - No coroutines - ```multireadrandom : 3.717 micros/op 269000 ops/sec 60.000 seconds 16140024 operations; 139.6 MB/s (16140024 of 16140024 found)``` Using coroutines - ```multireadrandom : 4.146 micros/op 241204 ops/sec 60.000 seconds 14472296 operations; 125.1 MB/s (14472296 of 14472296 found)``` 4. CPU bound multi-threaded (16 threads) with ~4 key overlap/file - No coroutines - ```multireadrandom : 4.534 micros/op 3528792 ops/sec 60.000 seconds 211728728 operations; 1830.7 MB/s (12737024 of 12737024 found) ``` Using coroutines - ```multireadrandom : 4.872 micros/op 3283812 ops/sec 60.000 seconds 197030096 operations; 1703.6 MB/s (12548032 of 12548032 found) ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/9968 Reviewed By: akankshamahajan15 Differential Revision: D36348563 Pulled By: anand1976 fbshipit-source-id: c0ce85a505fd26ebfbb09786cbd7f25202038696 |
3 years ago |
Peter Dillinger | 280b9f371a |
Fix auto_prefix_mode performance with partitioned filters (#10012)
Summary: Essentially refactored the RangeMayExist implementation in FullFilterBlockReader to FilterBlockReaderCommon so that it applies to partitioned filters as well. (The function is not called for the block-based filter case.) RangeMayExist is essentially a series of checks around a possible PrefixMayExist, and I'm confident those checks should be the same for partitioned as for full filters. (I think it's likely that bugs remain in those checks, but this change is overall a simplifying one.) Added auto_prefix_mode support to db_bench Other small fixes as well Fixes https://github.com/facebook/rocksdb/issues/10003 Pull Request resolved: https://github.com/facebook/rocksdb/pull/10012 Test Plan: Expanded unit test that uses statistics to check for filter optimization, fails without the production code changes here Performance: populate two DBs with ``` TEST_TMPDIR=/dev/shm/rocksdb_nonpartitioned ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=30000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=8 TEST_TMPDIR=/dev/shm/rocksdb_partitioned ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=30000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=8 -partition_index_and_filters ``` Observe no measurable change in non-partitioned performance ``` TEST_TMPDIR=/dev/shm/rocksdb_nonpartitioned ./db_bench -benchmarks=seekrandom[-X1000] -num=10000000 -readonly -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=8 -auto_prefix_mode -cache_index_and_filter_blocks=1 -cache_size=1000000000 -duration 20 ``` Before: seekrandom [AVG 15 runs] : 11798 (± 331) ops/sec After: seekrandom [AVG 15 runs] : 11724 (± 315) ops/sec Observe big improvement with partitioned (also supported by bloom use statistics) ``` TEST_TMPDIR=/dev/shm/rocksdb_partitioned ./db_bench -benchmarks=seekrandom[-X1000] -num=10000000 -readonly -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=8 -partition_index_and_filters -auto_prefix_mode -cache_index_and_filter_blocks=1 -cache_size=1000000000 -duration 20 ``` Before: seekrandom [AVG 12 runs] : 2942 (± 57) ops/sec After: seekrandom [AVG 12 runs] : 7489 (± 184) ops/sec Reviewed By: siying Differential Revision: D36469796 Pulled By: pdillinger fbshipit-source-id: bcf1e2a68d347b32adb2b27384f945434e7a266d |
3 years ago |
Jay Zhuang | c6d326d3d7 |
Track SST unique id in MANIFEST and verify (#9990)
Summary: Start tracking SST unique id in MANIFEST, which is used to verify with SST properties to make sure the SST file is not overwritten or misplaced. A DB option `try_verify_sst_unique_id` is introduced to enable/disable the verification, if enabled, it opens all SST files during DB-open to read the unique_id from table properties (default is false), so it's recommended to use it with `max_open_files = -1` to pre-open the files. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9990 Test Plan: unittests, format-compatible test, mini-crash Reviewed By: anand1976 Differential Revision: D36381863 Pulled By: jay-zhuang fbshipit-source-id: 89ea2eb6b35ed3e80ead9c724eb096083eaba63f |
3 years ago |
Hui Xiao | dde774db64 |
Mark old reserve* option deprecated (#10016)
Summary: **Context/Summary:** https://github.com/facebook/rocksdb/pull/9926 removed inefficient `reserve*` option API but forgot to mark them deprecated in `block_based_table_type_info` for compatible table format. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10016 Test Plan: build-format-compatible Reviewed By: pdillinger Differential Revision: D36484247 Pulled By: hx235 fbshipit-source-id: c41b90cc99fb7ab7098934052f0af7290b221f98 |
3 years ago |
gitbw95 | 4da34b97ee |
Set Read rate limiter priority dynamically and pass it to FS (#9996)
Summary: ### Context: Background compactions and flush generate large reads and writes, and can be long running, especially for universal compaction. In some cases, this can impact foreground reads and writes by users. ### Solution User, Flush, and Compaction reads share some code path. For this task, we update the rate_limiter_priority in ReadOptions for code paths (e.g. FindTable (mainly in BlockBasedTable::Open()) and various iterators), and eventually update the rate_limiter_priority in IOOptions for FSRandomAccessFile. **This PR is for the Read path.** The **Read:** dynamic priority for different state are listed as follows: | State | Normal | Delayed | Stalled | | ----- | ------ | ------- | ------- | | Flush (verification read in BuildTable()) | IO_USER | IO_USER | IO_USER | | Compaction | IO_LOW | IO_USER | IO_USER | | User | User provided | User provided | User provided | We will respect the read_options that the user provided and will not set it. The only sst read for Flush is the verification read in BuildTable(). It claims to be "regard as user read". **Details** 1. Set read_options.rate_limiter_priority dynamically: - User: Do not update the read_options. Use the read_options that the user provided. - Compaction: Update read_options in CompactionJob::ProcessKeyValueCompaction(). - Flush: Update read_options in BuildTable(). 2. Pass the rate limiter priority to FSRandomAccessFile functions: - After calling the FindTable(), read_options is passed through GetTableReader(table_cache.cc), BlockBasedTableFactory::NewTableReader(block_based_table_factory.cc), and BlockBasedTable::Open(). The Open() needs some updates for the ReadOptions variable and the updates are also needed for the called functions, including PrefetchTail(), PrepareIOOptions(), ReadFooterFromFile(), ReadMetaIndexblock(), ReadPropertiesBlock(), PrefetchIndexAndFilterBlocks(), and ReadRangeDelBlock(). - In RandomAccessFileReader, the functions to be updated include Read(), MultiRead(), ReadAsync(), and Prefetch(). - Update the downstream functions of NewIndexIterator(), NewDataBlockIterator(), and BlockBasedTableIterator(). ### Test Plans Add unit tests. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9996 Reviewed By: anand1976 Differential Revision: D36452483 Pulled By: gitbw95 fbshipit-source-id: 60978204a4f849bb9261cb78d9bc1cb56d6008cf |
3 years ago |
Peter Dillinger | 0070680cfd |
Adjust public APIs to prefer 128-bit SST unique ID (#10009)
Summary: 128 bits should suffice almost always and for tracking in manifest. Note that this changes the output of sst_dump --show_properties to only show 128 bits. Also introduces InternalUniqueIdToHumanString for presenting internal IDs for debugging purposes. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10009 Test Plan: unit tests updated Reviewed By: jay-zhuang Differential Revision: D36458189 Pulled By: pdillinger fbshipit-source-id: 93ebc4a3b6f9c73ee154383a1f8b291a5d6bbef5 |
3 years ago |
Hui Xiao | 3573558ec5 |
Rewrite memory-charging feature's option API (#9926)
Summary: **Context:** Previous PR https://github.com/facebook/rocksdb/pull/9748, https://github.com/facebook/rocksdb/pull/9073, https://github.com/facebook/rocksdb/pull/8428 added separate flag for each charged memory area. Such API design is not scalable as we charge more and more memory areas. Also, we foresee an opportunity to consolidate this feature with other cache usage related features such as `cache_index_and_filter_blocks` using `CacheEntryRole`. Therefore we decided to consolidate all these flags with `CacheUsageOptions cache_usage_options` and this PR serves as the first step by consolidating memory-charging related flags. **Summary:** - Replaced old API reference with new ones, including making `kCompressionDictionaryBuildingBuffer` opt-out and added a unit test for that - Added missing db bench/stress test for some memory charging features - Renamed related test suite to indicate they are under the same theme of memory charging - Refactored a commonly used mocked cache component in memory charging related tests to reduce code duplication - Replaced the phrases "memory tracking" / "cache reservation" (other than CacheReservationManager-related ones) with "memory charging" for standard description of this feature. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9926 Test Plan: - New unit test for opt-out `kCompressionDictionaryBuildingBuffer` `TEST_F(ChargeCompressionDictionaryBuildingBufferTest, Basic)` - New unit test for option validation/sanitization `TEST_F(CacheUsageOptionsOverridesTest, SanitizeAndValidateOptions)` - CI - db bench (in case querying new options introduces regression) **+0.5% micros/op**: `TEST_TMPDIR=/dev/shm/testdb ./db_bench -benchmarks=fillseq -db=$TEST_TMPDIR -charge_compression_dictionary_building_buffer=1(remove this for comparison) -compression_max_dict_bytes=10000 -disable_auto_compactions=1 -write_buffer_size=100000 -num=4000000 | egrep 'fillseq'` #-run | (pre-PR) avg micros/op | std micros/op | (post-PR) micros/op | std micros/op | change (%) -- | -- | -- | -- | -- | -- 10 | 3.9711 | 0.264408 | 3.9914 | 0.254563 | 0.5111933721 20 | 3.83905 | 0.0664488 | 3.8251 | 0.0695456 | **-0.3633711465** 40 | 3.86625 | 0.136669 | 3.8867 | 0.143765 | **0.5289363078** - db_stress: `python3 tools/db_crashtest.py blackbox -charge_compression_dictionary_building_buffer=1 -charge_filter_construction=1 -charge_table_reader=1 -cache_size=1` killed as normal Reviewed By: ajkr Differential Revision: D36054712 Pulled By: hx235 fbshipit-source-id: d406e90f5e0c5ea4dbcb585a484ad9302d4302af |
3 years ago |
mrambacher | b11ff347b4 |
Use STATIC_AVOID_DESTRUCTION for static objects with non-trivial destructors (#9958)
Summary: Changed the static objects that had non-trivial destructors to use the STATIC_AVOID_DESTRUCTION construct. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9958 Reviewed By: pdillinger Differential Revision: D36442982 Pulled By: mrambacher fbshipit-source-id: 029d47b1374d30d198bfede369a4c0ae7a4eb519 |
3 years ago |
mrambacher | 204a42ca97 |
Added GetFactoryCount/Names/Types to ObjectRegistry (#9358)
Summary: These methods allow for more thorough testing of the ObjectRegistry and Customizable infrastructure in a simpler manner. With this change, the Customizable tests can now check what factories are registered and attempt to create each of them in a systematic fashion. With this change, I think all of the factories registered with the ObjectRegistry/CreateFromString are now tested via the customizable_test classes. Note that there were a few other minor changes. There was a "posix://*" register with the ObjectRegistry which was missed during the PatternEntry conversion -- these changes found that. The nickname and default names for the FileSystem classes was also inverted. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9358 Reviewed By: pdillinger Differential Revision: D33433542 Pulled By: mrambacher fbshipit-source-id: 9a32da74e6620745b4eeffb2712be70eeeadfa7e |
3 years ago |
sdong | c4cd8e1acc |
Fix a bug handling multiget index I/O error. (#9993)
Summary: In one path of BlockBasedTable::MultiGet(), Next() is directly called after calling Seek() against the index iterator. This might cause crash if an I/O error happens in Seek(). The bug is discovered in crash test. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9993 Test Plan: See existing CI tests pass. Reviewed By: anand1976 Differential Revision: D36381758 fbshipit-source-id: a11e0aa48dcee168c2554c33b532646ffdb68877 |
3 years ago |
sdong | 736a7b5433 |
Remove own ToString() (#9955)
Summary: ToString() is created as some platform doesn't support std::to_string(). However, we've already used std::to_string() by mistake for 16 months (in db/db_info_dumper.cc). This commit just remove ToString(). Pull Request resolved: https://github.com/facebook/rocksdb/pull/9955 Test Plan: Watch CI tests Reviewed By: riversand963 Differential Revision: D36176799 fbshipit-source-id: bdb6dcd0e3a3ab96a1ac810f5d0188f684064471 |
3 years ago |
sdong | 49628c9a83 |
Use std::numeric_limits<> (#9954)
Summary: Right now we still don't fully use std::numeric_limits but use a macro, mainly for supporting VS 2013. Right now we only support VS 2017 and up so it is not a problem. The code comment claims that MinGW still needs it. We don't have a CI running MinGW so it's hard to validate. since we now require C++17, it's hard to imagine MinGW would still build RocksDB but doesn't support std::numeric_limits<>. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9954 Test Plan: See CI Runs. Reviewed By: riversand963 Differential Revision: D36173954 fbshipit-source-id: a35a73af17cdcae20e258cdef57fcf29a50b49e0 |
3 years ago |
Xinyu Zeng | 8b74cea7fe |
Reduce comparator objects init cost in BlockIter (#9611)
Summary: This PR solves the problem discussed in https://github.com/facebook/rocksdb/issues/7149. By storing the pointer of InternalKeyComparator as icmp_ in BlockIter, the object size remains the same. And for each call to CompareCurrentKey, there is no need to create Comparator objects. One can use icmp_ directly or use the "user_comparator" from the icmp_. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9611 Test Plan: with https://github.com/facebook/rocksdb/issues/9903, ``` $ TEST_TMPDIR=/dev/shm python3.6 ../benchmark/tools/compare.py benchmarks ./db_basic_bench ../rocksdb-pr9611/db_basic_bench --benchmark_filter=DBGet/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:0/negative_query:0/enable_filter:0/mmap:1/iterations:262144/threads:1 --benchmark_repetitions=50 ... Comparing ./db_basic_bench to ../rocksdb-pr9611/db_basic_bench Benchmark Time CPU Time Old Time New CPU Old CPU New ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ... DBGet/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:0/negative_query:0/enable_filter:0/mmap:1/iterations:262144/threads:1_pvalue 0.0001 0.0001 U Test, Repetitions: 50 vs 50 DBGet/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:0/negative_query:0/enable_filter:0/mmap:1/iterations:262144/threads:1_mean -0.0483 -0.0483 3924 3734 3924 3734 DBGet/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:0/negative_query:0/enable_filter:0/mmap:1/iterations:262144/threads:1_median -0.0713 -0.0713 3971 3687 3970 3687 DBGet/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:0/negative_query:0/enable_filter:0/mmap:1/iterations:262144/threads:1_stddev -0.0342 -0.0344 225 217 225 217 DBGet/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:0/negative_query:0/enable_filter:0/mmap:1/iterations:262144/threads:1_cv +0.0148 +0.0146 0 0 0 0 OVERALL_GEOMEAN -0.0483 -0.0483 0 0 0 0 ``` Reviewed By: akankshamahajan15 Differential Revision: D35882037 Pulled By: ajkr fbshipit-source-id: 9e5337bbad8f1239dff7aa9f6549020d599bfcdf |
3 years ago |
Peter Dillinger | 9d0cae7104 |
Eliminate unnecessary (slow) block cache Ref()ing in MultiGet (#9899)
Summary: When MultiGet() determines that multiple query keys can be served by examining the same data block in block cache (one Lookup()), each PinnableSlice referring to data in that data block needs to hold on to the block in cache so that they can be released at arbitrary times by the API user. Historically this is accomplished with extra calls to Ref() on the Handle from Lookup(), with each PinnableSlice cleanup calling Release() on the Handle, but this creates extra contention on the block cache for the extra Ref()s and Release()es, especially because they hit the same cache shard repeatedly. In the case of merge operands (possibly more cases?), the problem was compounded by doing an extra Ref()+eventual Release() for each merge operand for a key reusing a block (which could be the same key!), rather than one Ref() per key. (Note: the non-shared case with `biter` was already one per key.) This change optimizes MultiGet not to rely on these extra, contentious Ref()+Release() calls by instead, in the shared block case, wrapping the cache Release() cleanup in a refcounted object referenced by the PinnableSlices, such that after the last wrapped reference is released, the cache entry is Release()ed. Relaxed atomic refcounts should be much faster than mutex-guarded Ref() and Release(), and much less prone to a performance cliff when MultiGet() does a lot of block sharing. Note that I did not use std::shared_ptr, because that would require an extra indirection object (shared_ptr itself new/delete) in order to associate a ref increment/decrement with a Cleanable cleanup entry. (If I assumed it was the size of two pointers, I could do some hackery to make it work without the extra indirection, but that's too fragile.) Some details: * Fixed (removed) extra block cache tracing entries in cases of cache entry reuse in MultiGet, but it's likely that in some other cases traces are missing (XXX comment inserted) * Moved existing implementations for cleanable.h from iterator.cc to new cleanable.cc * Improved API comments on Cleanable * Added a public SharedCleanablePtr class to cleanable.h in case others could benefit from the same pattern (potentially many Cleanables and/or smart pointers referencing a shared Cleanable) * Add a typedef for MultiGetContext::Mask * Some variable renaming for clarity Pull Request resolved: https://github.com/facebook/rocksdb/pull/9899 Test Plan: Added unit tests for SharedCleanablePtr. Greatly enhanced ability of existing tests to detect cache use-after-free. * Release PinnableSlices from MultiGet as they are read rather than in bulk (in db_test_util wrapper). * In ASAN build, default to using a trivially small LRUCache for block_cache so that entries are immediately erased when unreferenced. (Updated two tests that depend on caching.) New ASAN testsuite running time seems OK to me. If I introduce a bug into my implementation where we skip the shared cleanups on block reuse, ASAN detects the bug in `db_basic_test *MultiGet*`. If I remove either of the above testing enhancements, the bug is not detected. Consider for follow-up work: manipulate or randomize ordering of PinnableSlice use and release from MultiGet db_test_util wrapper. But in typical cases, natural ordering gives pretty good functional coverage. Performance test: In the extreme (but possible) case of MultiGetting the same or adjacent keys in a batch, throughput can improve by an order of magnitude. `./db_bench -benchmarks=multireadrandom -db=/dev/shm/testdb -readonly -num=5 -duration=10 -threads=20 -multiread_batched -batch_size=200` Before ops/sec, num=5: 1,384,394 Before ops/sec, num=500: 6,423,720 After ops/sec, num=500: 10,658,794 After ops/sec, num=5: 16,027,257 Also note that previously, with high parallelism, having query keys concentrated in a single block was worse than spreading them out a bit. Now concentrated in a single block is faster than spread out, which is hopefully consistent with natural expectation. Random query performance: with num=1000000, over 999 x 10s runs running before & after simultaneously (each -threads=12): Before: multireadrandom [AVG 999 runs] : 1088699 (± 7344) ops/sec; 120.4 (± 0.8 ) MB/sec After: multireadrandom [AVG 999 runs] : 1090402 (± 7230) ops/sec; 120.6 (± 0.8 ) MB/sec Possibly better, possibly in the noise. Reviewed By: anand1976 Differential Revision: D35907003 Pulled By: pdillinger fbshipit-source-id: bbd244d703649a8ca12d476f2d03853ed9d1a17e |
3 years ago |
Akanksha Mahajan | 3653029dda |
Add stats related to async prefetching (#9845)
Summary: Add stats PREFETCHED_BYTES_DISCARDED and POLL_WAIT_MICROS. PREFETCHED_BYTES_DISCARDED records number of prefetched bytes discarded by FilePrefetchBuffer. POLL_WAIT_MICROS records the time taken by underling file_system Poll API. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9845 Test Plan: Update existing tests Reviewed By: anand1976 Differential Revision: D35909694 Pulled By: akankshamahajan15 fbshipit-source-id: e009ef940bb9ed72c9446f5529095caabb8a1e36 |
3 years ago |
Hui Xiao | a5063c8931 |
Fix issue of opening too many files in BlockBasedTableReaderCapMemoryTest.CapMemoryUsageUnderCacheCapacity (#9869)
Summary: **Context:** Unit test for https://github.com/facebook/rocksdb/pull/9748 keeps opening new files to see whether the new feature is able to correctly constrain the opening based on block cache capacity. However, the unit test has two places written inefficiently that can lead to opening too many new files relative to underlying operating system/file system constraint, even before hitting the block cache capacity: (1) [opened_table_reader_num < 2 * max_table_reader_num](https://github.com/facebook/rocksdb/pull/9748/files?show-viewed-files=true&file-filters%5B%5D=#diff-ec9f5353e317df71093094734ba29193b94a998f0f9c9af924e4c99692195eeaR438), which can leads to 1200 + open files because of (2) below (2) NewLRUCache(6 * CacheReservationManagerImpl<CacheEntryRole::kBlockBasedTableReader>::GetDummyEntrySize()) in [here](https://github.com/facebook/rocksdb/pull/9748/files?show-viewed-files=true&file-filters%5B%5D=#diff-ec9f5353e317df71093094734ba29193b94a998f0f9c9af924e4c99692195eeaR364) Therefore we see CI failures like this on machine with a strict open file limit ~1000 (see the "table_1021" naming in following error msg) https://app.circleci.com/pipelines/github/facebook/rocksdb/12886/workflows/75524682-3fa4-41ee-9a61-81827b51d99b/jobs/345270 ``` fs_->NewWritableFile(path, foptions, &file, nullptr) IO error: While open a file for appending: /dev/shm/rocksdb.Jedwt/run-block_based_table_reader_test-CapMemoryUsageUnderCacheCapacity-BlockBasedTableReaderCapMemoryTest.CapMemoryUsageUnderCacheCapacity-0/block_based_table_reader_test_1668910_829492452552920927/**table_1021**: Too many open files ``` **Summary:** - Revised the test more efficiently on the above 2 places, including using 1.1 instead 2 in the threshold and lowering down the block cache capacity a bit - Renamed some variables for clarity Pull Request resolved: https://github.com/facebook/rocksdb/pull/9869 Test Plan: - Manual inspection of max opened table reader in all test case, which is around ~389 - Circle CI to see if error is gone Reviewed By: ajkr Differential Revision: D35752655 Pulled By: hx235 fbshipit-source-id: 8a0953d39d561babfa4257b8ed8550bb21b04839 |
3 years ago |
Akanksha Mahajan | 0c7f455f85 |
Make initial auto readahead_size configurable (#9836)
Summary: Make initial auto readahead_size configurable Pull Request resolved: https://github.com/facebook/rocksdb/pull/9836 Test Plan: Added new unit test Ran regression: Without change: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` With this change: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Set seed to 1649895440554504 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.2 Date: Wed Apr 13 17:17:20 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] ... finished 100 ops seekrandom : 476892.488 micros/op 2 ops/sec; 344.6 MB/s (252 of 252 found) ``` Reviewed By: anand1976 Differential Revision: D35632815 Pulled By: akankshamahajan15 fbshipit-source-id: c8057a88f9294c9d03b1d434b03affe02f74d796 |
3 years ago |
Peter Dillinger | efd035164b |
Meta-internal folly integration with F14FastMap (#9546)
Summary: Especially after updating to C++17, I don't see a compelling case for *requiring* any folly components in RocksDB. I was able to purge the existing hard dependencies, and it can be quite difficult to strip out non-trivial components from folly for use in RocksDB. (The prospect of doing that on F14 has changed my mind on the best approach here.) But this change creates an optional integration where we can plug in components from folly at compile time, starting here with F14FastMap to replace std::unordered_map when possible (probably no public APIs for example). I have replaced the biggest CPU users of std::unordered_map with compile-time pluggable UnorderedMap which will use F14FastMap when USE_FOLLY is set. USE_FOLLY is always set in the Meta-internal buck build, and a simulation of that is in the Makefile for public CI testing. A full folly build is not needed, but checking out the full folly repo is much simpler for getting the dependency, and anything else we might want to optionally integrate in the future. Some picky details: * I don't think the distributed mutex stuff is actually used, so it was easy to remove. * I implemented an alternative to `folly::constexpr_log2` (which is much easier in C++17 than C++11) so that I could pull out the hard dependencies on `ConstexprMath.h` * I had to add noexcept move constructors/operators to some types to make F14's complainUnlessNothrowMoveAndDestroy check happy, and I added a macro to make that easier in some common cases. * Updated Meta-internal buck build to use folly F14Map (always) No updates to HISTORY.md nor INSTALL.md as this is not (yet?) considered a production integration for open source users. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9546 Test Plan: CircleCI tests updated so that a couple of them use folly. Most internal unit & stress/crash tests updated to use Meta-internal latest folly. (Note: they should probably use buck but they currently use Makefile.) Example performance improvement: when filter partitions are pinned in cache, they are tracked by PartitionedFilterBlockReader::filter_map_ and we can build a test that exercises that heavily. Build DB with ``` TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=30000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters ``` and test with (simultaneous runs with & without folly, ~20 times each to see convergence) ``` TEST_TMPDIR=/dev/shm/rocksdb ./db_bench_folly -readonly -use_existing_db -benchmarks=readrandom -num=10000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters -duration=40 -pin_l0_filter_and_index_blocks_in_cache ``` Average ops/s no folly: 26229.2 Average ops/s with folly: 26853.3 (+2.4%) Reviewed By: ajkr Differential Revision: D34181736 Pulled By: pdillinger fbshipit-source-id: ffa6ad5104c2880321d8a1aa7187e00ab0d02e94 |
3 years ago |
Akanksha Mahajan | 63e68a4e77 |
Enable async prefetching for ReadOptions.readahead_size (#9827)
Summary: Currently async prefetching is enabled for implicit internal auto readahead in FilePrefetchBuffer if `ReadOptions.async_io` is set. This PR enables async prefetching for `ReadOptions.readahead_size` when `ReadOptions.async_io` is set true. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9827 Test Plan: Update unit test Reviewed By: anand1976 Differential Revision: D35552129 Pulled By: akankshamahajan15 fbshipit-source-id: d9f9a96672852a591375a21eef15355cf3289f5c |
3 years ago |
Hui Xiao | f337542948 |
Fix a bug of TEST_SetRandomTableProperties due to non-zero padding between fields in TableProperties struct (#9812)
Summary: Context: https://github.com/facebook/rocksdb/pull/9748#discussion_r843134214 reveals an issue with TEST_SetRandomTableProperties when non-zero padding is used between the last string field and first non-string field in TableProperties. Fixed by https://github.com/facebook/rocksdb/pull/9748#discussion_r843244375 Pull Request resolved: https://github.com/facebook/rocksdb/pull/9812 Test Plan: No production code changes and rely on existing CI Reviewed By: ajkr Differential Revision: D35423680 Pulled By: hx235 fbshipit-source-id: fd855eef3d32771bb79c65bd7012ab8bb3c400ab |
3 years ago |
Peter Dillinger | 8ce7cea93f |
Tests for filter compatibility (#9773)
Summary: This change adds two unit tests that would each catch the regression fixed in https://github.com/facebook/rocksdb/issues/9736 * TableMetaIndexKeys - detects any churn in metaindex block keys generated by SST files using standard db_test_util configurations. * BloomFilterCompatibility - this detects if any common built-in FilterPolicy configurations fail to read filters generated by another. (The regression bug caused NewRibbonFilterPolicy not to read filters from NewBloomFilterPolicy and vice-versa.) This replaces some previous tests that didn't really appear to be testing much of anything except basic data correctness, which doesn't tell you a filter is being used. Light refactoring in meta_blocks.cc/h to support inspecting metaindex keys. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9773 Test Plan: this is the test. Verified that 7.0.2 fails both tests and 7.0.3 passes. With backporting for intentional API changes in 7.0, 6.29 also passes. Reviewed By: ajkr Differential Revision: D35236248 Pulled By: pdillinger fbshipit-source-id: 493dfe9ad7e27524bf7c6c1af8a4b8c31bc6ef5a |
3 years ago |
Hui Xiao | 49623f9c8e |
Account memory of big memory users in BlockBasedTable in global memory limit (#9748)
Summary: **Context:** Through heap profiling, we discovered that `BlockBasedTableReader` objects can accumulate and lead to high memory usage (e.g, `max_open_file = -1`). These memories are currently not saved, not tracked, not constrained and not cache evict-able. As a first step to improve this, similar to https://github.com/facebook/rocksdb/pull/8428, this PR is to track an estimate of `BlockBasedTableReader` object's memory in block cache and fail future creation if the memory usage exceeds the available space of cache at the time of creation. **Summary:** - Approximate big memory users (`BlockBasedTable::Rep` and `TableProperties` )' memory usage in addition to the existing estimated ones (filter block/index block/un-compression dictionary) - Charge all of these memory usages to block cache on `BlockBasedTable::Open()` and release them on `~BlockBasedTable()` as there is no memory usage fluctuation of concern in between - Refactor on CacheReservationManager (and its call-sites) to add concurrent support for BlockBasedTable used in this PR. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9748 Test Plan: - New unit tests - db bench: `OpenDb` : **-0.52% in ms** - Setup `./db_bench -benchmarks=fillseq -db=/dev/shm/testdb -disable_auto_compactions=1 -write_buffer_size=1048576` - Repeated run with pre-change w/o feature and post-change with feature, benchmark `OpenDb`: `./db_bench -benchmarks=readrandom -use_existing_db=1 -db=/dev/shm/testdb -reserve_table_reader_memory=true (remove this when running w/o feature) -file_opening_threads=3 -open_files=-1 -report_open_timing=true| egrep 'OpenDb:'` #-run | (feature-off) avg milliseconds | std milliseconds | (feature-on) avg milliseconds | std milliseconds | change (%) -- | -- | -- | -- | -- | -- 10 | 11.4018 | 5.95173 | 9.47788 | 1.57538 | -16.87382694 20 | 9.23746 | 0.841053 | 9.32377 | 1.14074 | 0.9343477536 40 | 9.0876 | 0.671129 | 9.35053 | 1.11713 | 2.893283155 80 | 9.72514 | 2.28459 | 9.52013 | 1.0894 | -2.108041632 160 | 9.74677 | 0.991234 | 9.84743 | 1.73396 | 1.032752389 320 | 10.7297 | 5.11555 | 10.547 | 1.97692 | **-1.70275031** 640 | 11.7092 | 2.36565 | 11.7869 | 2.69377 | **0.6635807741** - db bench on write with cost to cache in WriteBufferManager (just in case this PR's CRM refactoring accidentally slows down anything in WBM) : `fillseq` : **+0.54% in micros/op** `./db_bench -benchmarks=fillseq -db=/dev/shm/testdb -disable_auto_compactions=1 -cost_write_buffer_to_cache=true -write_buffer_size=10000000000 | egrep 'fillseq'` #-run | (pre-PR) avg micros/op | std micros/op | (post-PR) avg micros/op | std micros/op | change (%) -- | -- | -- | -- | -- | -- 10 | 6.15 | 0.260187 | 6.289 | 0.371192 | 2.260162602 20 | 7.28025 | 0.465402 | 7.37255 | 0.451256 | 1.267813605 40 | 7.06312 | 0.490654 | 7.13803 | 0.478676 | **1.060579461** 80 | 7.14035 | 0.972831 | 7.14196 | 0.92971 | **0.02254791432** - filter bench: `bloom filter`: **-0.78% in ms/key** - ` ./filter_bench -impl=2 -quick -reserve_table_builder_memory=true | grep 'Build avg'` #-run | (pre-PR) avg ns/key | std ns/key | (post-PR) ns/key | std ns/key | change (%) -- | -- | -- | -- | -- | -- 10 | 26.4369 | 0.442182 | 26.3273 | 0.422919 | **-0.4145720565** 20 | 26.4451 | 0.592787 | 26.1419 | 0.62451 | **-1.1465262** - Crash test `python3 tools/db_crashtest.py blackbox --reserve_table_reader_memory=1 --cache_size=1` killed as normal Reviewed By: ajkr Differential Revision: D35136549 Pulled By: hx235 fbshipit-source-id: 146978858d0f900f43f4eb09bfd3e83195e3be28 |
3 years ago |
Akanksha Mahajan | 36bc3da97f |
Fix segfault in FilePrefetchBuffer with async_io enabled (#9777)
Summary: If FilePrefetchBuffer object is destroyed and then later Poll() calls callback on object which has been destroyed, it gives segfault on accessing destroyed object. It was caught after adding unit tests that tests Posix implementation of ReadAsync and Poll APIs. This PR also updates and fixes existing IOURing tests which were not running locally because RocksDbIOUringEnable function wasn't defined and IOUring was disabled for those tests Pull Request resolved: https://github.com/facebook/rocksdb/pull/9777 Test Plan: Added new unit test Reviewed By: anand1976 Differential Revision: D35254002 Pulled By: akankshamahajan15 fbshipit-source-id: 68e80054ffb14ae25c255920ebc6548ca5f130a1 |
3 years ago |