Summary:
Currently it's easy to use a ton of memory with many small OptimisticTransactionDB instances, because each one by default allocates a million mutexes (40 bytes each on my compiler) for validating transactions. It even puts a lot of pressure on the allocator by allocating each one individually!
In this change:
* Create a new object and option that enables sharing these buckets of mutexes between instances. This is generally good for load balancing potential contention as various DBs become hotter or colder with txn writes. About the only cases where this sharing wouldn't make sense (e.g. each DB usually written by one thread) are cases that would be better off with OccValidationPolicy::kValidateSerial which doesn't use the buckets anyway.
* Allocate the mutexes in a contiguous array, for efficiency
* Add an option to ensure the mutexes are cache-aligned. In several other places we use cache-aligned mutexes but OptimisticTransactionDB historically does not. It should be a space-time trade-off the user can choose.
* Provide some visibility into the memory used by the mutex buckets with an ApproximateMemoryUsage() function (also used in unit testing)
* Share code with other users of "striped" mutexes, appropriate refactoring for customization & efficiency (e.g. using FastRange instead of modulus)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11439
Test Plan: unit tests added. Ran sized-up versions of stress test in unit test, including a before-and-after performance test showing no consistent difference. (NOTE: OptimisticTransactionDB not currently covered by db_stress!)
Reviewed By: ltamasi
Differential Revision: D45796393
Pulled By: pdillinger
fbshipit-source-id: ae2b3a26ad91ceeec15debcdc63ff48df6736a54
Summary:
add option `block_protection_bytes_per_key` and implementation for block per key-value checksum. The main changes are
1. checksum construction and verification in block.cc/h
2. pass the option `block_protection_bytes_per_key` around (mainly for methods defined in table_cache.h)
3. unit tests/crash test updates
Tests:
* Added unit tests
* Crash test: `python3 tools/db_crashtest.py blackbox --simple --block_protection_bytes_per_key=1 --write_buffer_size=1048576`
Follow up (maybe as a separate PR): make sure corruption status returned from BlockIters are correctly handled.
Performance:
Turning on block per KV protection has a non-trivial negative impact on read performance and costs additional memory.
For memory, each block includes additional 24 bytes for checksum-related states beside checksum itself. For CPU, I set up a DB of size ~1.2GB with 5M keys (32 bytes key and 200 bytes value) which compacts to ~5 SST files (target file size 256 MB) in L6 without compression. I tested readrandom performance with various block cache size (to mimic various cache hit rates):
```
SETUP
make OPTIMIZE_LEVEL="-O3" USE_LTO=1 DEBUG_LEVEL=0 -j32 db_bench
./db_bench -benchmarks=fillseq,compact0,waitforcompaction,compact,waitforcompaction -write_buffer_size=33554432 -level_compaction_dynamic_level_bytes=true -max_background_jobs=8 -target_file_size_base=268435456 --num=5000000 --key_size=32 --value_size=200 --compression_type=none
BENCHMARK
./db_bench --use_existing_db -benchmarks=readtocache,readrandom[-X10] --num=5000000 --key_size=32 --disable_auto_compactions --reads=1000000 --block_protection_bytes_per_key=[0|1] --cache_size=$CACHESIZE
The readrandom ops/sec looks like the following:
Block cache size: 2GB 1.2GB * 0.9 1.2GB * 0.8 1.2GB * 0.5 8MB
Main 240805 223604 198176 161653 139040
PR prot_bytes=0 238691 226693 200127 161082 141153
PR prot_bytes=1 214983 193199 178532 137013 108211
prot_bytes=1 vs -10% -15% -10.8% -15% -23%
prot_bytes=0
```
The benchmark has a lot of variance, but there was a 5% to 25% regression in this benchmark with different cache hit rates.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11287
Reviewed By: ajkr
Differential Revision: D43970708
Pulled By: cbi42
fbshipit-source-id: ef98d898b71779846fa74212b9ec9e08b7183940
Summary:
**Context:**
The existing stat rocksdb.sst.read.micros does not reflect each of compaction and flush cases but aggregate them, which is not so helpful for us to understand IO read behavior of each of them.
**Summary**
- Update `StopWatch` and `RandomAccessFileReader` to record `rocksdb.sst.read.micros` and `rocksdb.file.{flush/compaction}.read.micros`
- Fixed the default histogram in `RandomAccessFileReader`
- New field `ReadOptions/IOOptions::io_activity`; Pass `ReadOptions` through paths under db open, flush and compaction to where we can prepare `IOOptions` and pass it to `RandomAccessFileReader`
- Use `thread_status_util` for assertion in `DbStressFSWrapper` for continuous testing on we are passing correct `io_activity` under db open, flush and compaction
Pull Request resolved: https://github.com/facebook/rocksdb/pull/11288
Test Plan:
- **Stress test**
- **Db bench 1: rocksdb.sst.read.micros COUNT ≈ sum of rocksdb.file.read.flush.micros's and rocksdb.file.read.compaction.micros's.** (without blob)
- May not be exactly the same due to `HistogramStat::Add` only guarantees atomic not accuracy across threads.
```
./db_bench -db=/dev/shm/testdb/ -statistics=true -benchmarks="fillseq" -key_size=32 -value_size=512 -num=50000 -write_buffer_size=655 -target_file_size_base=655 -disable_auto_compactions=false -compression_type=none -bloom_bits=3 (-use_plain_table=1 -prefix_size=10)
```
```
// BlockBasedTable
rocksdb.sst.read.micros P50 : 2.009374 P95 : 4.968548 P99 : 8.110362 P100 : 43.000000 COUNT : 40456 SUM : 114805
rocksdb.file.read.flush.micros P50 : 1.871841 P95 : 3.872407 P99 : 5.540541 P100 : 43.000000 COUNT : 2250 SUM : 6116
rocksdb.file.read.compaction.micros P50 : 2.023109 P95 : 5.029149 P99 : 8.196910 P100 : 26.000000 COUNT : 38206 SUM : 108689
// PlainTable
Does not apply
```
- **Db bench 2: performance**
**Read**
SETUP: db with 900 files
```
./db_bench -db=/dev/shm/testdb/ -benchmarks="fillseq" -key_size=32 -value_size=512 -num=50000 -write_buffer_size=655 -disable_auto_compactions=true -target_file_size_base=655 -compression_type=none
```run till convergence
```
./db_bench -seed=1678564177044286 -use_existing_db=true -db=/dev/shm/testdb -benchmarks=readrandom[-X60] -statistics=true -num=1000000 -disable_auto_compactions=true -compression_type=none -bloom_bits=3
```
Pre-change
`readrandom [AVG 60 runs] : 21568 (± 248) ops/sec`
Post-change (no regression, -0.3%)
`readrandom [AVG 60 runs] : 21486 (± 236) ops/sec`
**Compaction/Flush**run till convergence
```
./db_bench -db=/dev/shm/testdb2/ -seed=1678564177044286 -benchmarks="fillseq[-X60]" -key_size=32 -value_size=512 -num=50000 -write_buffer_size=655 -disable_auto_compactions=false -target_file_size_base=655 -compression_type=none
rocksdb.sst.read.micros COUNT : 33820
rocksdb.sst.read.flush.micros COUNT : 1800
rocksdb.sst.read.compaction.micros COUNT : 32020
```
Pre-change
`fillseq [AVG 46 runs] : 1391 (± 214) ops/sec; 0.7 (± 0.1) MB/sec`
Post-change (no regression, ~-0.4%)
`fillseq [AVG 46 runs] : 1385 (± 216) ops/sec; 0.7 (± 0.1) MB/sec`
Reviewed By: ajkr
Differential Revision: D44007011
Pulled By: hx235
fbshipit-source-id: a54c89e4846dfc9a135389edf3f3eedfea257132
Summary:
This is several refactorings bundled into one to avoid having to incrementally re-modify uses of Cache several times. Overall, there are breaking changes to Cache class, and it becomes more of low-level interface for implementing caches, especially block cache. New internal APIs make using Cache cleaner than before, and more insulated from block cache evolution. Hopefully, this is the last really big block cache refactoring, because of rather effectively decoupling the implementations from the uses. This change also removes the EXPERIMENTAL designation on the SecondaryCache support in Cache. It seems reasonably mature at this point but still subject to change/evolution (as I warn in the API docs for Cache).
The high-level motivation for this refactoring is to minimize code duplication / compounding complexity in adding SecondaryCache support to HyperClockCache (in a later PR). Other benefits listed below.
* static_cast lines of code +29 -35 (net removed 6)
* reinterpret_cast lines of code +6 -32 (net removed 26)
## cache.h and secondary_cache.h
* Always use CacheItemHelper with entries instead of just a Deleter. There are several motivations / justifications:
* Simpler for implementations to deal with just one Insert and one Lookup.
* Simpler and more efficient implementation because we don't have to track which entries are using helpers and which are using deleters
* Gets rid of hack to classify cache entries by their deleter. Instead, the CacheItemHelper includes a CacheEntryRole. This simplifies a lot of code (cache_entry_roles.h almost eliminated). Fixes https://github.com/facebook/rocksdb/issues/9428.
* Makes it trivial to adjust SecondaryCache behavior based on kind of block (e.g. don't re-compress filter blocks).
* It is arguably less convenient for many direct users of Cache, but direct users of Cache are now rare with introduction of typed_cache.h (below).
* I considered and rejected an alternative approach in which we reduce customizability by assuming each secondary cache compatible value starts with a Slice referencing the uncompressed block contents (already true or mostly true), but we apparently intend to stack secondary caches. Saving an entry from a compressed secondary to a lower tier requires custom handling offered by SaveToCallback, etc.
* Make CreateCallback part of the helper and introduce CreateContext to work with it (alternative to https://github.com/facebook/rocksdb/issues/10562). This cleans up the interface while still allowing context to be provided for loading/parsing values into primary cache. This model works for async lookup in BlockBasedTable reader (reader owns a CreateContext) under the assumption that it always waits on secondary cache operations to finish. (Otherwise, the CreateContext could be destroyed while async operation depending on it continues.) This likely contributes most to the observed performance improvement because it saves an std::function backed by a heap allocation.
* Use char* for serialized data, e.g. in SaveToCallback, where void* was confusingly used. (We use `char*` for serialized byte data all over RocksDB, with many advantages over `void*`. `memcpy` etc. are legacy APIs that should not be mimicked.)
* Add a type alias Cache::ObjectPtr = void*, so that we can better indicate the intent of the void* when it is to be the object associated with a Cache entry. Related: started (but did not complete) a refactoring to move away from "value" of a cache entry toward "object" or "obj". (It is confusing to call Cache a key-value store (like DB) when it is really storing arbitrary in-memory objects, not byte strings.)
* Remove unnecessary key param from DeleterFn. This is good for efficiency in HyperClockCache, which does not directly store the cache key in memory. (Alternative to https://github.com/facebook/rocksdb/issues/10774)
* Add allocator to Cache DeleterFn. This is a kind of future-proofing change in case we get more serious about using the Cache allocator for memory tracked by the Cache. Right now, only the uncompressed block contents are allocated using the allocator, and a pointer to that allocator is saved as part of the cached object so that the deleter can use it. (See CacheAllocationPtr.) If in the future we are able to "flatten out" our Cache objects some more, it would be good not to have to track the allocator as part of each object.
* Removes legacy `ApplyToAllCacheEntries` and changes `ApplyToAllEntries` signature for Deleter->CacheItemHelper change.
## typed_cache.h
Adds various "typed" interfaces to the Cache as internal APIs, so that most uses of Cache can use simple type safe code without casting and without explicit deleters, etc. Almost all of the non-test, non-glue code uses of Cache have been migrated. (Follow-up work: CompressedSecondaryCache deserves deeper attention to migrate.) This change expands RocksDB's internal usage of metaprogramming and SFINAE (https://en.cppreference.com/w/cpp/language/sfinae).
The existing usages of Cache are divided up at a high level into these new interfaces. See updated existing uses of Cache for examples of how these are used.
* PlaceholderCacheInterface - Used for making cache reservations, with entries that have a charge but no value.
* BasicTypedCacheInterface<TValue> - Used for primary cache storage of objects of type TValue, which can be cleaned up with std::default_delete<TValue>. The role is provided by TValue::kCacheEntryRole or given in an optional template parameter.
* FullTypedCacheInterface<TValue, TCreateContext> - Used for secondary cache compatible storage of objects of type TValue. In addition to BasicTypedCacheInterface constraints, we require TValue::ContentSlice() to return persistable data. This simplifies usage for the normal case of simple secondary cache compatibility (can give you a Slice to the data already in memory). In addition to TCreateContext performing the role of Cache::CreateContext, it is also expected to provide a factory function for creating TValue.
* For each of these, there's a "Shared" version (e.g. FullTypedSharedCacheInterface) that holds a shared_ptr to the Cache, rather than assuming external ownership by holding only a raw `Cache*`.
These interfaces introduce specific handle types for each interface instantiation, so that it's easy to see what kind of object is controlled by a handle. (Ultimately, this might not be worth the extra complexity, but it seems OK so far.)
Note: I attempted to make the cache 'charge' automatically inferred from the cache object type, such as by expecting an ApproximateMemoryUsage() function, but this is not so clean because there are cases where we need to compute the charge ahead of time and don't want to re-compute it.
## block_cache.h
This header is essentially the replacement for the old block_like_traits.h. It includes various things to support block cache access with typed_cache.h for block-based table.
## block_based_table_reader.cc
Before this change, accessing the block cache here was an awkward mix of static polymorphism (template TBlocklike) and switch-case on a dynamic BlockType value. This change mostly unifies on static polymorphism, relying on minor hacks in block_cache.h to distinguish variants of Block. We still check BlockType in some places (especially for stats, which could be improved in follow-up work) but at least the BlockType is a static constant from the template parameter. (No more awkward partial redundancy between static and dynamic info.) This likely contributes to the overall performance improvement, but hasn't been tested in isolation.
The other key source of simplification here is a more unified system of creating block cache objects: for directly populating from primary cache and for promotion from secondary cache. Both use BlockCreateContext, for context and for factory functions.
## block_based_table_builder.cc, cache_dump_load_impl.cc
Before this change, warming caches was super ugly code. Both of these source files had switch statements to basically transition from the dynamic BlockType world to the static TBlocklike world. None of that mess is needed anymore as there's a new, untyped WarmInCache function that handles all the details just as promotion from SecondaryCache would. (Fixes `TODO akanksha: Dedup below code` in block_based_table_builder.cc.)
## Everything else
Mostly just updating Cache users to use new typed APIs when reasonably possible, or changed Cache APIs when not.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10975
Test Plan:
tests updated
Performance test setup similar to https://github.com/facebook/rocksdb/issues/10626 (by cache size, LRUCache when not "hyper" for HyperClockCache):
34MB 1thread base.hyper -> kops/s: 0.745 io_bytes/op: 2.52504e+06 miss_ratio: 0.140906 max_rss_mb: 76.4844
34MB 1thread new.hyper -> kops/s: 0.751 io_bytes/op: 2.5123e+06 miss_ratio: 0.140161 max_rss_mb: 79.3594
34MB 1thread base -> kops/s: 0.254 io_bytes/op: 1.36073e+07 miss_ratio: 0.918818 max_rss_mb: 45.9297
34MB 1thread new -> kops/s: 0.252 io_bytes/op: 1.36157e+07 miss_ratio: 0.918999 max_rss_mb: 44.1523
34MB 32thread base.hyper -> kops/s: 7.272 io_bytes/op: 2.88323e+06 miss_ratio: 0.162532 max_rss_mb: 516.602
34MB 32thread new.hyper -> kops/s: 7.214 io_bytes/op: 2.99046e+06 miss_ratio: 0.168818 max_rss_mb: 518.293
34MB 32thread base -> kops/s: 3.528 io_bytes/op: 1.35722e+07 miss_ratio: 0.914691 max_rss_mb: 264.926
34MB 32thread new -> kops/s: 3.604 io_bytes/op: 1.35744e+07 miss_ratio: 0.915054 max_rss_mb: 264.488
233MB 1thread base.hyper -> kops/s: 53.909 io_bytes/op: 2552.35 miss_ratio: 0.0440566 max_rss_mb: 241.984
233MB 1thread new.hyper -> kops/s: 62.792 io_bytes/op: 2549.79 miss_ratio: 0.044043 max_rss_mb: 241.922
233MB 1thread base -> kops/s: 1.197 io_bytes/op: 2.75173e+06 miss_ratio: 0.103093 max_rss_mb: 241.559
233MB 1thread new -> kops/s: 1.199 io_bytes/op: 2.73723e+06 miss_ratio: 0.10305 max_rss_mb: 240.93
233MB 32thread base.hyper -> kops/s: 1298.69 io_bytes/op: 2539.12 miss_ratio: 0.0440307 max_rss_mb: 371.418
233MB 32thread new.hyper -> kops/s: 1421.35 io_bytes/op: 2538.75 miss_ratio: 0.0440307 max_rss_mb: 347.273
233MB 32thread base -> kops/s: 9.693 io_bytes/op: 2.77304e+06 miss_ratio: 0.103745 max_rss_mb: 569.691
233MB 32thread new -> kops/s: 9.75 io_bytes/op: 2.77559e+06 miss_ratio: 0.103798 max_rss_mb: 552.82
1597MB 1thread base.hyper -> kops/s: 58.607 io_bytes/op: 1449.14 miss_ratio: 0.0249324 max_rss_mb: 1583.55
1597MB 1thread new.hyper -> kops/s: 69.6 io_bytes/op: 1434.89 miss_ratio: 0.0247167 max_rss_mb: 1584.02
1597MB 1thread base -> kops/s: 60.478 io_bytes/op: 1421.28 miss_ratio: 0.024452 max_rss_mb: 1589.45
1597MB 1thread new -> kops/s: 63.973 io_bytes/op: 1416.07 miss_ratio: 0.0243766 max_rss_mb: 1589.24
1597MB 32thread base.hyper -> kops/s: 1436.2 io_bytes/op: 1357.93 miss_ratio: 0.0235353 max_rss_mb: 1692.92
1597MB 32thread new.hyper -> kops/s: 1605.03 io_bytes/op: 1358.04 miss_ratio: 0.023538 max_rss_mb: 1702.78
1597MB 32thread base -> kops/s: 280.059 io_bytes/op: 1350.34 miss_ratio: 0.023289 max_rss_mb: 1675.36
1597MB 32thread new -> kops/s: 283.125 io_bytes/op: 1351.05 miss_ratio: 0.0232797 max_rss_mb: 1703.83
Almost uniformly improving over base revision, especially for hot paths with HyperClockCache, up to 12% higher throughput seen (1597MB, 32thread, hyper). The improvement for that is likely coming from much simplified code for providing context for secondary cache promotion (CreateCallback/CreateContext), and possibly from less branching in block_based_table_reader. And likely a small improvement from not reconstituting key for DeleterFn.
Reviewed By: anand1976
Differential Revision: D42417818
Pulled By: pdillinger
fbshipit-source-id: f86bfdd584dce27c028b151ba56818ad14f7a432
Summary:
Although we've been tracking SST unique IDs in the DB manifest
unconditionally, checking has been opt-in and with an extra pass at DB::Open
time. This changes the behavior of `verify_sst_unique_id_in_manifest` to
check unique ID against manifest every time an SST file is opened through
table cache (normal DB operations), replacing the explicit pass over files
at DB::Open time. This change also enables the option by default and
removes the "EXPERIMENTAL" designation.
One possible criticism is that the option no longer ensures the integrity
of a DB at Open time. This is far from an all-or-nothing issue. Verifying
the IDs of all SST files hardly ensures all the data in the DB is readable.
(VerifyChecksum is supposed to do that.) Also, with
max_open_files=-1 (default, extremely common), all SST files are
opened at DB::Open time anyway.
Implementation details:
* `VerifySstUniqueIdInManifest()` functions are the extra/explicit pass
that is now removed.
* Unit tests that manipulate/corrupt table properties have to opt out of
this check, because that corrupts the "actual" unique id. (And even for
testing we don't currently have a mechanism to set "no unique id"
in the in-memory file metadata for new files.)
* A lot of other unit test churn relates to (a) default checking on, and
(b) checking on SST open even without DB::Open (e.g. on flush)
* Use `FileMetaData` for more `TableCache` operations (in place of
`FileDescriptor`) so that we have access to the unique_id whenever
we might need to open an SST file. **There is the possibility of
performance impact because we can no longer use the more
localized `fd` part of an `FdWithKeyRange` but instead follow the
`file_metadata` pointer. However, this change (possible regression)
is only done for `GetMemoryUsageByTableReaders`.**
* Removed a completely unnecessary constructor overload of
`TableReaderOptions`
Possible follow-up:
* Verification only happens when opening through table cache. Are there
more places where this should happen?
* Improve error message when there is a file size mismatch vs. manifest
(FIXME added in the appropriate place).
* I'm not sure there's a justification for `FileDescriptor` to be distinct from
`FileMetaData`.
* I'm skeptical that `FdWithKeyRange` really still makes sense for
optimizing some data locality by duplicating some data in memory, but I
could be wrong.
* An unnecessary overload of NewTableReader was recently added, in
the public API nonetheless (though unusable there). It should be cleaned
up to put most things under `TableReaderOptions`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10532
Test Plan:
updated unit tests
Performance test showing no significant difference (just noise I think):
`./db_bench -benchmarks=readwhilewriting[-X10] -num=3000000 -disable_wal=1 -bloom_bits=8 -write_buffer_size=1000000 -target_file_size_base=1000000`
Before: readwhilewriting [AVG 10 runs] : 68702 (± 6932) ops/sec
After: readwhilewriting [AVG 10 runs] : 68239 (± 7198) ops/sec
Reviewed By: jay-zhuang
Differential Revision: D38765551
Pulled By: pdillinger
fbshipit-source-id: a827a708155f12344ab2a5c16e7701c7636da4c2
Summary:
Delete range logic is moved from `DBIter` to `MergingIterator`, and `MergingIterator` will seek to the end of a range deletion if possible instead of scanning through each key and check with `RangeDelAggregator`.
With the invariant that a key in level L (consider memtable as the first level, each immutable and L0 as a separate level) has a larger sequence number than all keys in any level >L, a range tombstone `[start, end)` from level L covers all keys in its range in any level >L. This property motivates optimizations in iterator:
- in `Seek(target)`, if level L has a range tombstone `[start, end)` that covers `target.UserKey`, then for all levels > L, we can do Seek() on `end` instead of `target` to skip some range tombstone covered keys.
- in `Next()/Prev()`, if the current key is covered by a range tombstone `[start, end)` from level L, we can do `Seek` to `end` for all levels > L.
This PR implements the above optimizations in `MergingIterator`. As all range tombstone covered keys are now skipped in `MergingIterator`, the range tombstone logic is removed from `DBIter`. The idea in this PR is similar to https://github.com/facebook/rocksdb/issues/7317, but this PR leaves `InternalIterator` interface mostly unchanged. **Credit**: the cascading seek optimization and the sentinel key (discussed below) are inspired by [Pebble](https://github.com/cockroachdb/pebble/blob/master/merging_iter.go) and suggested by ajkr in https://github.com/facebook/rocksdb/issues/7317. The two optimizations are mostly implemented in `SeekImpl()/SeekForPrevImpl()` and `IsNextDeleted()/IsPrevDeleted()` in `merging_iterator.cc`. See comments for each method for more detail.
One notable change is that the minHeap/maxHeap used by `MergingIterator` now contains range tombstone end keys besides point key iterators. This helps to reduce the number of key comparisons. For example, for a range tombstone `[start, end)`, a `start` and an `end` `HeapItem` are inserted into the heap. When a `HeapItem` for range tombstone start key is popped from the minHeap, we know this range tombstone becomes "active" in the sense that, before the range tombstone's end key is popped from the minHeap, all the keys popped from this heap is covered by the range tombstone's internal key range `[start, end)`.
Another major change, *delete range sentinel key*, is made to `LevelIterator`. Before this PR, when all point keys in an SST file are iterated through in `MergingIterator`, a level iterator would advance to the next SST file in its level. In the case when an SST file has a range tombstone that covers keys beyond the SST file's last point key, advancing to the next SST file would lose this range tombstone. Consequently, `MergingIterator` could return keys that should have been deleted by some range tombstone. We prevent this by pretending that file boundaries in each SST file are sentinel keys. A `LevelIterator` now only advance the file iterator once the sentinel key is processed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10449
Test Plan:
- Added many unit tests in db_range_del_test
- Stress test: `./db_stress --readpercent=5 --prefixpercent=19 --writepercent=20 -delpercent=10 --iterpercent=44 --delrangepercent=2`
- Additional iterator stress test is added to verify against iterators against expected state: https://github.com/facebook/rocksdb/issues/10538. This is based on ajkr's previous attempt https://github.com/facebook/rocksdb/pull/5506#issuecomment-506021913.
```
python3 ./tools/db_crashtest.py blackbox --simple --write_buffer_size=524288 --target_file_size_base=524288 --max_bytes_for_level_base=2097152 --compression_type=none --max_background_compactions=8 --value_size_mult=33 --max_key=5000000 --interval=10 --duration=7200 --delrangepercent=3 --delpercent=9 --iterpercent=25 --writepercent=60 --readpercent=3 --prefixpercent=0 --num_iterations=1000 --range_deletion_width=100 --verify_iterator_with_expected_state_one_in=1
```
- Performance benchmark: I used a similar setup as in the blog [post](http://rocksdb.org/blog/2018/11/21/delete-range.html) that introduced DeleteRange, "a database with 5 million data keys, and 10000 range tombstones (ignoring those dropped during compaction) that were written in regular intervals after 4.5 million data keys were written". As expected, the performance with this PR depends on the range tombstone width.
```
# Setup:
TEST_TMPDIR=/dev/shm ./db_bench_main --benchmarks=fillrandom --writes=4500000 --num=5000000
TEST_TMPDIR=/dev/shm ./db_bench_main --benchmarks=overwrite --writes=500000 --num=5000000 --use_existing_db=true --writes_per_range_tombstone=50
# Scan entire DB
TEST_TMPDIR=/dev/shm ./db_bench_main --benchmarks=readseq[-X5] --use_existing_db=true --num=5000000 --disable_auto_compactions=true
# Short range scan (10 Next())
TEST_TMPDIR=/dev/shm/width-100/ ./db_bench_main --benchmarks=seekrandom[-X5] --use_existing_db=true --num=500000 --reads=100000 --seek_nexts=10 --disable_auto_compactions=true
# Long range scan(1000 Next())
TEST_TMPDIR=/dev/shm/width-100/ ./db_bench_main --benchmarks=seekrandom[-X5] --use_existing_db=true --num=500000 --reads=2500 --seek_nexts=1000 --disable_auto_compactions=true
```
Avg over of 10 runs (some slower tests had fews runs):
For the first column (tombstone), 0 means no range tombstone, 100-10000 means width of the 10k range tombstones, and 1 means there is a single range tombstone in the entire DB (width is 1000). The 1 tombstone case is to test regression when there's very few range tombstones in the DB, as no range tombstone is likely to take a different code path than with range tombstones.
- Scan entire DB
| tombstone width | Pre-PR ops/sec | Post-PR ops/sec | ±% |
| ------------- | ------------- | ------------- | ------------- |
| 0 range tombstone |2525600 (± 43564) |2486917 (± 33698) |-1.53% |
| 100 |1853835 (± 24736) |2073884 (± 32176) |+11.87% |
| 1000 |422415 (± 7466) |1115801 (± 22781) |+164.15% |
| 10000 |22384 (± 227) |227919 (± 6647) |+918.22% |
| 1 range tombstone |2176540 (± 39050) |2434954 (± 24563) |+11.87% |
- Short range scan
| tombstone width | Pre-PR ops/sec | Post-PR ops/sec | ±% |
| ------------- | ------------- | ------------- | ------------- |
| 0 range tombstone |35398 (± 533) |35338 (± 569) |-0.17% |
| 100 |28276 (± 664) |31684 (± 331) |+12.05% |
| 1000 |7637 (± 77) |25422 (± 277) |+232.88% |
| 10000 |1367 |28667 |+1997.07% |
| 1 range tombstone |32618 (± 581) |32748 (± 506) |+0.4% |
- Long range scan
| tombstone width | Pre-PR ops/sec | Post-PR ops/sec | ±% |
| ------------- | ------------- | ------------- | ------------- |
| 0 range tombstone |2262 (± 33) |2353 (± 20) |+4.02% |
| 100 |1696 (± 26) |1926 (± 18) |+13.56% |
| 1000 |410 (± 6) |1255 (± 29) |+206.1% |
| 10000 |25 |414 |+1556.0% |
| 1 range tombstone |1957 (± 30) |2185 (± 44) |+11.65% |
- Microbench does not show significant regression: https://gist.github.com/cbi42/59f280f85a59b678e7e5d8561e693b61
Reviewed By: ajkr
Differential Revision: D38450331
Pulled By: cbi42
fbshipit-source-id: b5ef12e8d8c289ed2e163ccdf277f5039b511fca
Summary:
The fix in https://github.com/facebook/rocksdb/issues/10513 was not complete w.r.t range deletion handling. It didn't handle the case where a file with a range tombstone covering a key also overlapped another key in the batch. In that case, ```mget_range``` would be non-empty. However, ```mget_range``` would only have the second key and, therefore, the first key would be skipped when iterating through the range tombstones in ```TableCache::MultiGet```.
Test plan -
1. Add a unit test
2. Run stress tests
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10534
Reviewed By: akankshamahajan15
Differential Revision: D38773880
Pulled By: anand1976
fbshipit-source-id: dae491dbe52e18bbce5179b77b63f20771a66c00
Summary:
This PR fixes 2 bugs introduced in https://github.com/facebook/rocksdb/issues/10432 -
1. If the bloom filter returned a negative result for all MultiGet keys in a file, the range tombstones in that file were being ignored, resulting in incorrect results if those tombstones covered a key in a higher level.
2. If all the keys in a file were filtered out in `TableCache::MultiGetFilter`, the table cache handle was not being released.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10513
Test Plan: Add a new unit test that fails without this fix
Reviewed By: akankshamahajan15
Differential Revision: D38548739
Pulled By: anand1976
fbshipit-source-id: a741a1e25d2e991d63f038100f126c2dc404a87c
Summary:
This PR is the first step in enhancing the coroutines MultiGet to be able to lookup a batch in parallel across levels. By having a separate TableReader function for probing the bloom filters, we can quickly figure out which overlapping keys from a batch are definitely not in the file and can move on to the next level.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10432
Reviewed By: akankshamahajan15
Differential Revision: D38245910
Pulled By: anand1976
fbshipit-source-id: 3d20db2350378c3fe6f086f0c7ba5ff01d7f04de
Summary:
Unit tests still haven't been fixed. Also need to add more tests. But I ran some simple fillrandom db_bench and the partitioning feels reasonable.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10393
Test Plan:
1. Make sure existing tests pass. This should cover some basic sub compaction logic to be correct and the partitioning result is reasonable;
2. Add a new unit test to ApproximateKeyAnchors()
3. Run some db_bench with max_subcompaction = 4 and watch the compaction is indeed partitioned evenly.
Reviewed By: jay-zhuang
Differential Revision: D38043783
fbshipit-source-id: 085008e0f85f9b7c5abff7800307618320efb19f
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
Summary:
Fixes a major performance regression in 6.26, where
extra CPU is spent in SliceTransform::AsString when reads involve
a prefix_extractor (Get, MultiGet, Seek). Common case performance
is now better than 6.25.
This change creates a "fast path" for verifying that the current prefix
extractor is unchanged and compatible with what was used to
generate a table file. This fast path detects the common case by
pointer comparison on the current prefix_extractor and a "known
good" prefix extractor (if applicable) that is saved at the time the
table reader is opened. The "known good" prefix extractor is saved
as another shared_ptr copy (in an existing field, however) to ensure
the pointer is not recycled.
When the prefix_extractor has changed to a different instance but
same compatible configuration (rare, odd), performance is still a
regression compared to 6.25, but this is likely acceptable because
of the oddity of such a case. The performance of incompatible
prefix_extractor is essentially unchanged.
Also fixed a minor case (ForwardIterator) where a prefix_extractor
could be used via a raw pointer after being freed as a shared_ptr,
if replaced via SetOptions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9407
Test Plan:
## Performance
Populate DB with `TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=10000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=12`
Running head-to-head comparisons simultaneously with `TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -use_existing_db -readonly -benchmarks=seekrandom -num=10000000 -duration=20 -disable_wal=1 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=12`
Below each is compared by ops/sec vs. baseline which is version 6.25 (multiple baseline runs because of variable machine load)
v6.26: 4833 vs. 6698 (<- major regression!)
v6.27: 4737 vs. 6397 (still)
New: 6704 vs. 6461 (better than baseline in common case)
Disabled fastpath: 4843 vs. 6389 (e.g. if prefix extractor instance changes but is still compatible)
Changed prefix size (no usable filter) in new: 787 vs. 5927
Changed prefix size (no usable filter) in new & baseline: 773 vs. 784
Reviewed By: mrambacher
Differential Revision: D33677812
Pulled By: pdillinger
fbshipit-source-id: 571d9711c461fb97f957378a061b7e7dbc4d6a76
Summary:
* Checksums are now checked on meta blocks unless specifically
suppressed or not applicable (e.g. plain table). (Was other way around.)
This means a number of cases that were not checking checksums now are,
including direct read TableProperties in Version::GetTableProperties
(fixed in meta_blocks ReadTableProperties), reading any block from
PersistentCache (fixed in BlockFetcher), read TableProperties in
SstFileDumper (ldb/sst_dump/BackupEngine) before table reader open,
maybe more.
* For that to work, I moved the global_seqno+TableProperties checksum
logic to the shared table/ code, because that is used by many utilies
such as SstFileDumper.
* Also for that to work, we have to know when we're dealing with a block
that has a checksum (trailer), so added that capability to Footer based
on magic number, and from there BlockFetcher.
* Knowledge of trailer presence has also fixed a problem where other
table formats were reading blocks including bytes for a non-existant
trailer--and awkwardly kind-of not using them, e.g. no shared code
checking checksums. (BlockFetcher compression type was populated
incorrectly.) Now we only read what is needed.
* Minimized code duplication and differing/incompatible/awkward
abstractions in meta_blocks.{cc,h} (e.g. SeekTo in metaindex block
without parsing block handle)
* Moved some meta block handling code from table_properties*.*
* Moved some code specific to block-based table from shared table/ code
to BlockBasedTable class. The checksum stuff means we can't completely
separate it, but things that don't need to be in shared table/ code
should not be.
* Use unique_ptr rather than raw ptr in more places. (Note: you can
std::move from unique_ptr to shared_ptr.)
Without enhancements to GetPropertiesOfAllTablesTest (see below),
net reduction of roughly 100 lines of code.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9163
Test Plan:
existing tests and
* Enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to verify that
checksums are now checked on direct read of table properties by TableCache
(new test would fail before this change)
* Also enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to test
putting table properties under old meta name
* Also generally enhanced that same test to actually test what it was
supposed to be testing already, by kicking things out of table cache when
we don't want them there.
Reviewed By: ajkr, mrambacher
Differential Revision: D32514757
Pulled By: pdillinger
fbshipit-source-id: 507964b9311d186ae8d1131182290cbd97a99fa9
Summary:
For tiered storage project, we need to know the block read count and read bytes of files with different temperature. Add FileIOByTemperature to IOStatsContext and collect the bytes read and read count from different temperature files through the RandomAccessFileReader.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8710
Test Plan: make check, add the testing cases
Reviewed By: siying
Differential Revision: D30582400
Pulled By: zhichao-cao
fbshipit-source-id: d83173de594374fc8404af5ce93a6a9be72c7141
Summary:
FileOptions has an implicit conversion from EnvOptions and some
internal APIs take `const FileOptions&` and save the reference, which is
counter to Google C++ guidelines,
> Avoid defining functions that require a const reference parameter to outlive the call, because const reference parameters bind to temporaries. Instead, find a way to eliminate the lifetime requirement (for example, by copying the parameter), or pass it by const pointer and document the lifetime and non-null requirements.
This is at least a problem for repair.cc, which passes an EnvOptions to
TableCache(), which would save a reference to the temporary copy as
FileOptions. This was unfortunately only caught as a side effect of
changes in https://github.com/facebook/rocksdb/issues/8544.
This change fixes the repair.cc case and updates the involved internal
APIs that save a reference to use `const FileOptions*` instead.
Unfortunately, I don't know how to get any of our sanitizers to reliably
report bugs like this, so I can't rule out more existing in our
codebase.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8571
Test Plan:
Test that issues seen with https://github.com/facebook/rocksdb/issues/8544 are fixed (can reproduce on
AWS EC2)
Reviewed By: ajkr
Differential Revision: D29943890
Pulled By: pdillinger
fbshipit-source-id: 95f9c5251548777b4dc994c1a083dd2add5799c9
Summary:
Currently, we either use the file system inode or a monotonically incrementing runtime ID as the block cache key prefix. However, if we use a monotonically incrementing runtime ID (in the case that the file system does not support inode id generation), in some cases, it cannot ensure uniqueness (e.g., we have secondary cache migrated from host to host). We use DbSessionID (20 bytes) + current file number (at most 10 bytes) as the new cache block key prefix when the secondary cache is enabled. So can accommodate scenarios such as transfer of cache state across hosts.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8360
Test Plan: add the test to lru_cache_test
Reviewed By: pdillinger
Differential Revision: D29006215
Pulled By: zhichao-cao
fbshipit-source-id: 6cff686b38d83904667a2bd39923cd030df16814
Summary:
The ImmutableCFOptions contained a bunch of fields that belonged to the ImmutableDBOptions. This change cleans that up by introducing an ImmutableOptions struct. Following the pattern of Options struct, this class inherits from the DB and CFOption structs (of the Immutable form).
Only one structural change (the ImmutableCFOptions::fs was changed to a shared_ptr from a raw one) is in this PR. All of the other changes involve moving the member variables from the ImmutableCFOptions into the ImmutableOptions and changing member variables or function parameters as required for compilation purposes.
Follow-on PRs may do a further clean-up of the code, such as renaming variables (such as "ImmutableOptions cf_options") and potentially eliminating un-needed function parameters (there is no longer a need to pass both an ImmutableDBOptions and an ImmutableOptions to a function).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8262
Reviewed By: pdillinger
Differential Revision: D28226540
Pulled By: mrambacher
fbshipit-source-id: 18ae71eadc879dedbe38b1eb8e6f9ff5c7147dbf
Summary:
Replace FSRandomAccessFile pointer with FSRandomAccessFilePtr
object in RandomAccessFileReader.
This new object wraps FSRandomAccessFile pointer.
Objective: If tracing is enabled, FSRandomAccessFile Ptr returns
FSRandomAccessFileTracingWrapper pointer that includes all necessary
information in IORecord and calls underlying FileSystem and invokes
IOTracer to dump that record in a binary file. If tracing is disabled
then, underlying FileSystem pointer is returned directly.
FSRandomAccessFilePtr wrapper class is added to bypass the FSRandomAccessFileWrapper when
tracing is disabled.
Test Plan: make check -j64
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7192
Reviewed By: anand1976
Differential Revision: D23356867
Pulled By: akankshamahajan15
fbshipit-source-id: 48f31168166a17a7444b40be44a9a9d4a5c7182c
Summary:
Previously, a `ReadOptions` object was stored in every `BlockBasedTableIterator`
and every `LevelIterator`. This redundancy consumes extra memory,
resulting in the `Arena` making more allocations, and iteration
observing worse cache performance.
This PR migrates callers of `NewInternalIterator()` and
`MakeInputIterator()` to provide a `ReadOptions` object guaranteed to
outlive the returned iterator. When the iterator's lifetime will be managed by the
user, this lifetime guarantee is achieved by storing the `ReadOptions`
value in `ArenaWrappedDBIter`. Then, sub-iterators of `NewInternalIterator()` and
`MakeInputIterator()` can hold a reference-to-const `ReadOptions`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7210
Test Plan:
- `make check` under ASAN and valgrind
- benchmark: on a DB with 2 L0 files and 3 L1+ levels, this PR reduced `Arena` allocation 4792 -> 4160 bytes.
Reviewed By: anand1976
Differential Revision: D22861323
Pulled By: ajkr
fbshipit-source-id: 54aebb3e89c872eeab0f5793b4b6e42878d093ce
Summary:
Current implementation of the ```read_options.deadline``` option only checks the deadline for random file reads during point lookups. This PR extends the checks to file opens, prefetches and preloads as part of table open.
The main changes are in the ```BlockBasedTable```, partitioned index and filter readers, and ```TableCache``` to take ReadOptions as an additional parameter. In ```BlockBasedTable::Open```, in order to retain existing behavior w.r.t checksum verification and block cache usage, we filter out most of the options in ```ReadOptions``` except ```deadline```. However, having the ```ReadOptions``` gives us more flexibility to honor other options like verify_checksums, fill_cache etc. in the future.
Additional changes in callsites due to function signature changes in ```NewTableReader()``` and ```FilePrefetchBuffer```.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6982
Test Plan: Add new unit tests in db_basic_test
Reviewed By: riversand963
Differential Revision: D22219515
Pulled By: anand1976
fbshipit-source-id: 8a3b92f4a889808013838603aa3ca35229cd501b
Summary:
Memory pinned by `pin_l0_filter_and_index_blocks_in_cache` needs to be predictable based on user config. This PR makes sure
we do not pin extra memory for large files generated by intra-L0 (see https://github.com/facebook/rocksdb/issues/6889).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6911
Test Plan: unit test
Reviewed By: siying
Differential Revision: D21835818
Pulled By: ajkr
fbshipit-source-id: a11a088549d06bed8aacc2548d266e5983f0ead4
Summary:
Context: Index type `kBinarySearchWithFirstKey` added the ability for sst file iterator to sometimes report a key from index without reading the corresponding data block. This is useful when sst blocks are cut at some meaningful boundaries (e.g. one block per key prefix), and many seeks land between blocks (e.g. for each prefix, the ranges of keys in different sst files are nearly disjoint, so a typical seek needs to read a data block from only one file even if all files have the prefix). But this added a new error condition, which rocksdb code was really not equipped to deal with: `InternalIterator::value()` may fail with an IO error or Status::Incomplete, but it's just a method returning a Slice, with no way to report error instead. Before this PR, this type of error wasn't handled at all (an empty slice was returned), and kBinarySearchWithFirstKey implementation was considered a prototype.
Now that we (LogDevice) have experimented with kBinarySearchWithFirstKey for a while and confirmed that it's really useful, this PR is adding the missing error handling.
It's a pretty inconvenient situation implementation-wise. The error needs to be reported from InternalIterator when trying to access value. But there are ~700 call sites of `InternalIterator::value()`, most of which either can't hit the error condition (because the iterator is reading from memtable or from index or something) or wouldn't benefit from the deferred loading of the value (e.g. compaction iterator that reads all values anyway). Adding error handling to all these call sites would needlessly bloat the code. So instead I made the deferred value loading optional: only the call sites that may use deferred loading have to call the new method `PrepareValue()` before calling `value()`. The feature is enabled with a new bool argument `allow_unprepared_value` to a bunch of methods that create iterators (it wouldn't make sense to put it in ReadOptions because it's completely internal to iterators, with virtually no user-visible effect). Lmk if you have better ideas.
Note that the deferred value loading only happens for *internal* iterators. The user-visible iterator (DBIter) always prepares the value before returning from Seek/Next/etc. We could go further and add an API to defer that value loading too, but that's most likely not useful for LogDevice, so it doesn't seem worth the complexity for now.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6621
Test Plan: make -j5 check . Will also deploy to some logdevice test clusters and look at stats.
Reviewed By: siying
Differential Revision: D20786930
Pulled By: al13n321
fbshipit-source-id: 6da77d918bad3780522e918f17f4d5513d3e99ee
Summary:
When dynamically linking two binaries together, different builds of RocksDB from two sources might cause errors. To provide a tool for user to solve the problem, the RocksDB namespace is changed to a flag which can be overridden in build time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6433
Test Plan: Build release, all and jtest. Try to build with ROCKSDB_NAMESPACE with another flag.
Differential Revision: D19977691
fbshipit-source-id: aa7f2d0972e1c31d75339ac48478f34f6cfcfb3e
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
Summary:
Further apply formatter to more recent commits.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5830
Test Plan: Run all existing tests.
Differential Revision: D17488031
fbshipit-source-id: 137458fd94d56dd271b8b40c522b03036943a2ab
Summary:
VersionSet::ApproximateSize doesn't need to create two separate index iterators and do binary search for each in BlockBasedTable. So BlockBasedTable::ApproximateSize was added that creates the iterator once and uses it to calculate the data size between start and end keys.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5693
Differential Revision: D16774056
Pulled By: elipoz
fbshipit-source-id: 53ce262e1a057788243bf30cd9b8aa6581df1a18
Summary:
Add a command in ldb so that users can print out tombstones in SST files.
In order to test the code, change the interface of LDBCommandRunner::RunCommand() so that it doesn't return from the program, but return the status code.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5615
Test Plan: Add a new unit test
Differential Revision: D16550326
fbshipit-source-id: 88ddfe6984bdcbb3a528abdd115089df09eba52e
Summary:
1. Avoid creating the iterator in order to call BlockBasedTable::ApproximateOffsetOf(). Instead, directly call into it.
2. Optimize BlockBasedTable::ApproximateOffsetOf() keeps the index block iterator in stack.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5613
Differential Revision: D16442660
Pulled By: elipoz
fbshipit-source-id: 9320be3e918c139b10e758cbbb684706d172e516
Summary:
RandomAccessFileReader.for_compaction_ doesn't seem to be used anymore. Remove it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5572
Test Plan: USE_CLANG=1 make all check -j
Differential Revision: D16286178
fbshipit-source-id: aa338049761033dfbe5e8b1707bbb0be2df5be7e
Summary:
This PR adds more callers for table readers. These information are only used for block cache analysis so that we can know which caller accesses a block.
1. It renames the BlockCacheLookupCaller to TableReaderCaller as passing the caller from upstream requires changes to table_reader.h and TableReaderCaller is a more appropriate name.
2. It adds more table reader callers in table/table_reader_caller.h, e.g., kCompactionRefill, kExternalSSTIngestion, and kBuildTable.
This PR is long as it requires modification of interfaces in table_reader.h, e.g., NewIterator.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5454
Test Plan: make clean && COMPILE_WITH_ASAN=1 make check -j32.
Differential Revision: D15819451
Pulled By: HaoyuHuang
fbshipit-source-id: b6caa704c8fb96ddd15b9a934b7e7ea87f88092d
Summary:
Currently the read-ahead logic for user reads and compaction reads go through different code paths where compaction reads create new table readers and use `ReadaheadRandomAccessFile`. This change is to unify read-ahead logic to use read-ahead in BlockBasedTableReader::InitDataBlock(). As a result of the change `ReadAheadRandomAccessFile` class and `new_table_reader_for_compaction_inputs` option will no longer be used.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5431
Test Plan:
make check
Here is the benchmarking - https://gist.github.com/vjnadimpalli/083cf423f7b6aa12dcdb14c858bc18a5
Differential Revision: D15772533
Pulled By: vjnadimpalli
fbshipit-source-id: b71dca710590471ede6fb37553388654e2e479b9
Summary:
This PR integrates the block cache tracer class into db_impl.cc.
db_impl.cc contains a member variable of AtomicBlockCacheTraceWriter class and passes its reference to the block_based_table_reader.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5433
Differential Revision: D15728016
Pulled By: HaoyuHuang
fbshipit-source-id: 23d5659e8c82d556833dcc1a5558aac8c1f7db71
Summary:
This PR introduces a new MultiGet() API, with the underlying implementation grouping keys based on SST file and batching lookups in a file. The reason for the new API is twofold - the definition allows callers to allocate storage for status and values on stack instead of std::vector, as well as return values as PinnableSlices in order to avoid copying, and it keeps the original MultiGet() implementation intact while we experiment with batching.
Batching is useful when there is some spatial locality to the keys being queries, as well as larger batch sizes. The main benefits are due to -
1. Fewer function calls, especially to BlockBasedTableReader::MultiGet() and FullFilterBlockReader::KeysMayMatch()
2. Bloom filter cachelines can be prefetched, hiding the cache miss latency
The next step is to optimize the binary searches in the level_storage_info, index blocks and data blocks, since we could reduce the number of key comparisons if the keys are relatively close to each other. The batching optimizations also need to be extended to other formats, such as PlainTable and filter formats. This also needs to be added to db_stress.
Benchmark results from db_bench for various batch size/locality of reference combinations are given below. Locality was simulated by offsetting the keys in a batch by a stride length. Each SST file is about 8.6MB uncompressed and key/value size is 16/100 uncompressed. To focus on the cpu benefit of batching, the runs were single threaded and bound to the same cpu to eliminate interference from other system events. The results show a 10-25% improvement in micros/op from smaller to larger batch sizes (4 - 32).
Batch Sizes
1 | 2 | 4 | 8 | 16 | 32
Random pattern (Stride length 0)
4.158 | 4.109 | 4.026 | 4.05 | 4.1 | 4.074 - Get
4.438 | 4.302 | 4.165 | 4.122 | 4.096 | 4.075 - MultiGet (no batching)
4.461 | 4.256 | 4.277 | 4.11 | 4.182 | 4.14 - MultiGet (w/ batching)
Good locality (Stride length 16)
4.048 | 3.659 | 3.248 | 2.99 | 2.84 | 2.753
4.429 | 3.728 | 3.406 | 3.053 | 2.911 | 2.781
4.452 | 3.45 | 2.833 | 2.451 | 2.233 | 2.135
Good locality (Stride length 256)
4.066 | 3.786 | 3.581 | 3.447 | 3.415 | 3.232
4.406 | 4.005 | 3.644 | 3.49 | 3.381 | 3.268
4.393 | 3.649 | 3.186 | 2.882 | 2.676 | 2.62
Medium locality (Stride length 4096)
4.012 | 3.922 | 3.768 | 3.61 | 3.582 | 3.555
4.364 | 4.057 | 3.791 | 3.65 | 3.57 | 3.465
4.479 | 3.758 | 3.316 | 3.077 | 2.959 | 2.891
dbbench command used (on a DB with 4 levels, 12 million keys)-
TEST_TMPDIR=/dev/shm numactl -C 10 ./db_bench.tmp -use_existing_db=true -benchmarks="readseq,multireadrandom" -write_buffer_size=4194304 -target_file_size_base=4194304 -max_bytes_for_level_base=16777216 -num=12000000 -reads=12000000 -duration=90 -threads=1 -compression_type=none -cache_size=4194304000 -batch_size=32 -disable_auto_compactions=true -bloom_bits=10 -cache_index_and_filter_blocks=true -pin_l0_filter_and_index_blocks_in_cache=true -multiread_batched=true -multiread_stride=4
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5011
Differential Revision: D14348703
Pulled By: anand1976
fbshipit-source-id: 774406dab3776d979c809522a67bedac6c17f84b
Summary:
Choose to preload some files if options.max_open_files != -1. This can slightly narrow the gap of performance between options.max_open_files is -1 and a large number. To avoid a significant regression to DB reopen speed if options.max_open_files != -1. Limit the files to preload in DB open time to 16.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3340
Differential Revision: D6686945
Pulled By: siying
fbshipit-source-id: 8ec11bbdb46e3d0cdee7b6ad5897a09c5a07869f
Summary:
Now that v2 is fully functional, the v1 aggregator is removed.
The v2 aggregator has been renamed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4778
Differential Revision: D13495930
Pulled By: abhimadan
fbshipit-source-id: 9d69500a60a283e79b6c4fa938fc68a8aa4d40d6
Summary:
The old RangeDelAggregator did expensive pre-processing work
to create a collapsed, binary-searchable representation of range
tombstones. With FragmentedRangeTombstoneIterator, much of this work is
now unnecessary. RangeDelAggregatorV2 takes advantage of this by seeking
in each iterator to find a covering tombstone in ShouldDelete, while
doing minimal work in AddTombstones. The old RangeDelAggregator is still
used during flush/compaction for now, though RangeDelAggregatorV2 will
support those uses in a future PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4649
Differential Revision: D13146964
Pulled By: abhimadan
fbshipit-source-id: be29a4c020fc440500c137216fcc1cf529571eb3
Summary:
Ran the following commands to recursively change all the files under RocksDB:
```
find . -type f -name "*.cc" -exec sed -i 's/ unique_ptr/ std::unique_ptr/g' {} +
find . -type f -name "*.cc" -exec sed -i 's/<unique_ptr/<std::unique_ptr/g' {} +
find . -type f -name "*.cc" -exec sed -i 's/ shared_ptr/ std::shared_ptr/g' {} +
find . -type f -name "*.cc" -exec sed -i 's/<shared_ptr/<std::shared_ptr/g' {} +
```
Running `make format` updated some formatting on the files touched.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4638
Differential Revision: D12934992
Pulled By: sagar0
fbshipit-source-id: 45a15d23c230cdd64c08f9c0243e5183934338a8
Summary:
To more accurately truncate range tombstones at SST boundaries,
we now represent them in RangeDelAggregator using InternalKeys, which
are end-key-exclusive as they were before this change.
During compaction, "atomic compaction unit boundaries" (the range of
keys contained in neighbouring and overlaping SSTs) are propagated down
to RangeDelAggregator to truncate range tombstones at those boundariies
instead. See https://github.com/facebook/rocksdb/pull/4432#discussion_r221072219 and https://github.com/facebook/rocksdb/pull/4432#discussion_r221138683
for motivating examples.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4432
Differential Revision: D10263952
Pulled By: abhimadan
fbshipit-source-id: 2fe85ff8a02b3a6a2de2edfe708012797a7bd579
Summary:
Given that index value is a BlockHandle, which is basically an <offset, size> pair we can apply delta encoding on the values. The first value at each index restart interval encoded the full BlockHandle but the rest encode only the size. Refer to IndexBlockIter::DecodeCurrentValue for the detail of the encoding. This reduces the index size which helps using the block cache more efficiently. The feature is enabled with using format_version 4.
The feature comes with a bit of cpu overhead which should be paid back by the higher cache hits due to smaller index block size.
Results with sysbench read-only using 4k blocks and using 16 index restart interval:
Format 2:
19585 rocksdb read-only range=100
Format 3:
19569 rocksdb read-only range=100
Format 4:
19352 rocksdb read-only range=100
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3983
Differential Revision: D8361343
Pulled By: maysamyabandeh
fbshipit-source-id: f882ee082322acac32b0072e2bdbb0b5f854e651
Summary:
Do not consider the range tombstone sentinel key as causing 2 adjacent
sstables in a level to overlap. When a range tombstone's end key is the
largest key in an sstable, the sstable's end key is so to a "sentinel"
value that is the smallest key in the next sstable with a sequence
number of kMaxSequenceNumber. This "sentinel" is guaranteed to not
overlap in internal-key space with the next sstable. Unfortunately,
GetOverlappingFiles uses user-keys to determine overlap and was thus
considering 2 adjacent sstables in a level to overlap if they were
separated by this sentinel key. This in turn would cause compactions to
be larger than necessary.
Note that this conflicts with
https://github.com/facebook/rocksdb/pull/2769 and cases
`DBRangeDelTest.CompactionTreatsSplitInputLevelDeletionAtomically` to
fail.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4050
Differential Revision: D8844423
Pulled By: ajkr
fbshipit-source-id: df3f9f1db8f4cff2bff77376b98b83c2ae1d155b
Summary:
https://github.com/facebook/rocksdb/pull/3881 fixed a bug where PinnableSlice pin mmap files which could be deleted with background compaction. This is however a non-issue for ReadOnlyDB when there is no compaction running and max_open_files is -1. This patch reenables the pinning feature for that case.
Closes https://github.com/facebook/rocksdb/pull/4053
Differential Revision: D8662546
Pulled By: maysamyabandeh
fbshipit-source-id: 402962602eb0f644e17822748332999c3af029fd
Summary:
Currently it is not possible to change bloom filter config without restart the db, which is causing a lot of operational complexity for users.
This PR aims to make it possible to dynamically change bloom filter config.
Closes https://github.com/facebook/rocksdb/pull/3601
Differential Revision: D7253114
Pulled By: miasantreble
fbshipit-source-id: f22595437d3e0b86c95918c484502de2ceca120c
Summary:
Allow users to rate limit background work based on read bytes, written bytes, or sum of read and written bytes. Support these by changing the RateLimiter API, so no additional options were needed.
Closes https://github.com/facebook/rocksdb/pull/2433
Differential Revision: D5216946
Pulled By: ajkr
fbshipit-source-id: aec57a8357dbb4bfde2003261094d786d94f724e
