Summary:
The patch adds a new BlobDB configuration option `blob_compaction_readahead_size`
that can be used to enable prefetching data from blob files during compaction.
This is important when using storage with higher latencies like HDDs or remote filesystems.
If enabled, prefetching is used for all cases when blobs are read during compaction,
namely garbage collection, compaction filters (when the existing value has to be read from
a blob file), and `Merge` (when the value of the base `Put` is stored in a blob file).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9187
Test Plan: Ran `make check` and the stress/crash test.
Reviewed By: riversand963
Differential Revision: D32565512
Pulled By: ltamasi
fbshipit-source-id: 87be9cebc3aa01cc227bec6b5f64d827b8164f5d
Summary:
Allow compaction_job_test, db_io_failure_test, dbformat_test, deletefile_test, and fault_injection_test to use a custom Env object. Also move ```RegisterCustomObjects``` declaration to a header file to simplify things.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9087
Test Plan: Run manually using "buck test rocksdb/src:compaction_job_test_fbcode" etc.
Reviewed By: riversand963
Differential Revision: D32007222
Pulled By: anand1976
fbshipit-source-id: 99af58559e25bf61563dfa95dc46e31fa7375792
Summary:
The current BlobDB garbage collection logic works by relocating the valid
blobs from the oldest blob files as they are encountered during compaction,
and cleaning up blob files once they contain nothing but garbage. However,
with sufficiently skewed workloads, it is theoretically possible to end up in a
situation when few or no compactions get scheduled for the SST files that contain
references to the oldest blob files, which can lead to increased space amp due
to the lack of GC.
In order to efficiently handle such workloads, the patch adds a new BlobDB
configuration option called `blob_garbage_collection_force_threshold`,
which signals to BlobDB to schedule targeted compactions for the SST files
that keep alive the oldest batch of blob files if the overall ratio of garbage in
the given blob files meets the threshold *and* all the given blob files are
eligible for GC based on `blob_garbage_collection_age_cutoff`. (For example,
if the new option is set to 0.9, targeted compactions will get scheduled if the
sum of garbage bytes meets or exceeds 90% of the sum of total bytes in the
oldest blob files, assuming all affected blob files are below the age-based cutoff.)
The net result of these targeted compactions is that the valid blobs in the oldest
blob files are relocated and the oldest blob files themselves cleaned up (since
*all* SST files that rely on them get compacted away).
These targeted compactions are similar to periodic compactions in the sense
that they force certain SST files that otherwise would not get picked up to undergo
compaction and also in the sense that instead of merging files from multiple levels,
they target a single file. (Note: such compactions might still include neighboring files
from the same level due to the need of having a "clean cut" boundary but they never
include any files from any other level.)
This functionality is currently only supported with the leveled compaction style
and is inactive by default (since the default value is set to 1.0, i.e. 100%).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8994
Test Plan: Ran `make check` and tested using `db_bench` and the stress/crash tests.
Reviewed By: riversand963
Differential Revision: D31489850
Pulled By: ltamasi
fbshipit-source-id: 44057d511726a0e2a03c5d9313d7511b3f0c4eab
Summary:
Made SystemClock into a Customizable class, complete with CreateFromString.
Cleaned up some of the existing SystemClock implementations that were redundant (NoSleep was the same as the internal one for MockEnv).
Changed MockEnv construction to allow Clock to be passed to the Memory/MockFileSystem.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8636
Reviewed By: zhichao-cao
Differential Revision: D30483360
Pulled By: mrambacher
fbshipit-source-id: cd0e3a876c39f8c98fe13374c06e8edbd5b9f2a1
Summary:
This PR does the following:
-> Makes the MemTableRepFactory into a Customizable class and creatable/configurable via CreateFromString
-> Makes the existing implementations compatible with configurations
-> Moves the "SpecialRepFactory" test class into testutil, accessible via the ObjectRegistry or a NewSpecial API
New tests were added to validate the functionality and all existing tests pass. db_bench and memtablerep_bench were hand-tested to verify the functionality in those tools.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8419
Reviewed By: zhichao-cao
Differential Revision: D29558961
Pulled By: mrambacher
fbshipit-source-id: 81b7229636e4e649a0c914e73ac7b0f8454c931c
Summary:
- Fix issue with OptionType::Vector when the nested item is a Customizable with no names
- Fix issue with OptionType::Vector to appropriately wrap the elements in a Vector;
- Fix an issue with nested Customizable object with a null immutable object still appearing in the mutable options;
- Fix/Add tests for null/empty customizable objects
- Move the RegisterTestObjects from customizable_test into testutil.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8566
Reviewed By: zhichao-cao
Differential Revision: D30303724
Pulled By: mrambacher
fbshipit-source-id: 33fa8ea2a3b663210cb356da05e64aab7585b1b5
Summary:
- Added CreateFromString method to Env and FilesSystem to replace LoadEnv/Load. This method/signature is a precursor to making these classes extend Customizable.
- Added CreateFromSystem to Env. This method standardizes creating an Env from the environment variables. Previously, some places would check TEST_ENV_URI and others would also check TEST_FS_URI. Now the code is more command/standardized.
- Added CreateFromFlags to Env. These method allows Env to be create from string options (such as GFLAGS options) in a more standard way.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8174
Reviewed By: zhichao-cao
Differential Revision: D28999603
Pulled By: mrambacher
fbshipit-source-id: 88e6911e7e91f908458a7fe10a20e93ecbc275fb
Summary:
Introduces and uses a SystemClock class to RocksDB. This class contains the time-related functions of an Env and these functions can be redirected from the Env to the SystemClock.
Many of the places that used an Env (Timer, PerfStepTimer, RepeatableThread, RateLimiter, WriteController) for time-related functions have been changed to use SystemClock instead. There are likely more places that can be changed, but this is a start to show what can/should be done. Over time it would be nice to migrate most (if not all) of the uses of the time functions from the Env to the SystemClock.
There are several Env classes that implement these functions. Most of these have not been converted yet to SystemClock implementations; that will come in a subsequent PR. It would be good to unify many of the Mock Timer implementations, so that they behave similarly and be tested similarly (some override Sleep, some use a MockSleep, etc).
Additionally, this change will allow new methods to be introduced to the SystemClock (like https://github.com/facebook/rocksdb/issues/7101 WaitFor) in a consistent manner across a smaller number of classes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7858
Reviewed By: pdillinger
Differential Revision: D26006406
Pulled By: mrambacher
fbshipit-source-id: ed10a8abbdab7ff2e23d69d85bd25b3e7e899e90
Summary:
This PR does the following:
-> Creates a WinFileSystem class. This class is the Windows equivalent of the PosixFileSystem and will be used on Windows systems.
-> Introduces a CustomEnv class. A CustomEnv is an Env that takes a FileSystem as constructor argument. I believe there will only ever be two implementations of this class (PosixEnv and WinEnv). There is still a CustomEnvWrapper class that takes an Env and a FileSystem and wraps the Env calls with the input Env but uses the FileSystem for the FileSystem calls
-> Eliminates the public uses of the LegacyFileSystemWrapper.
With this change in place, there are effectively the following patterns of Env:
- "Base Env classes" (PosixEnv, WinEnv). These classes implement the core Env functions (e.g. Threads) and have a hard-coded input FileSystem. These classes inherit from CompositeEnv, implement the core Env functions (threads) and delegate the FileSystem-like calls to the input file system.
- Wrapped Composite Env classes (MemEnv). These classes take in an Env and a FileSystem. The core env functions are re-directed to the wrapped env. The file system calls are redirected to the input file system
- Legacy Wrapped Env classes. These classes take in an Env input (but no FileSystem). The core env functions are re-directed to the wrapped env. A "Legacy File System" is created using this env and the file system calls directed to the env itself.
With these changes in place, the PosixEnv becomes a singleton -- there is only ever one created. Any other use of the PosixEnv is via another wrapped env. This cleans up some of the issues with the env construction and destruction.
Additionally, there were places in the code that required had an Env when they required a FileSystem. Many of these places would wrap the Env with a LegacyFileSystemWrapper instead of using the env->GetFileSystem(). These places were changed, thereby removing layers of additional redirection (LegacyFileSystem --> Env --> Env::FileSystem).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7703
Reviewed By: zhichao-cao
Differential Revision: D25762190
Pulled By: anand1976
fbshipit-source-id: 1a088e97fc916f28ac69c149cd1dcad0ab31704b
Summary:
Change the StringEnv and related classes to be based on FileSystem APIs rather than the corresponding Env ones. The StringSink and StringSource classes were changed to be based on the corresponding FS file classes.
Part of a cleanup to use the newer interfaces. This change also eliminates some of the casts/wrappers to LegacyFile classes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7786
Reviewed By: jay-zhuang
Differential Revision: D25761460
Pulled By: anand1976
fbshipit-source-id: 428ae8e32b3db97dbeeca08c9d3bb0d9d4d3a38f
Summary:
This patch simply adds a couple of options that will enable users to
configure garbage collection when using the integrated BlobDB
implementation. The actual GC logic will be added in a separate step.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7661
Test Plan: `make check`
Reviewed By: riversand963
Differential Revision: D24906544
Pulled By: ltamasi
fbshipit-source-id: ee0e056a712a4b4475cd90de8b27d969bd61b7e1
Summary:
This PR does a few things:
1. The MockFileSystem class was split out from the MockEnv. This change would theoretically allow a MockFileSystem to be used by other Environments as well (if we created a means of constructing one). The MockFileSystem implements a FileSystem in its entirety and does not rely on any Wrapper implementation.
2. Make the RocksDB test suite work when MOCK_ENV=1 and ENCRYPTED_ENV=1 are set. To accomplish this, a few things were needed:
- The tests that tried to use the "wrong" environment (Env::Default() instead of env_) were updated
- The MockFileSystem was changed to support the features it was missing or mishandled (such as recursively deleting files in a directory or supporting renaming of a directory).
3. Updated the test framework to have a ROCKSDB_GTEST_SKIP macro. This can be used to flag tests that are skipped. Currently, this defaults to doing nothing (marks the test as SUCCESS) but will mark the tests as SKIPPED when RocksDB is upgraded to a version of gtest that supports this (gtest-1.10).
I have run a full "make check" with MEM_ENV, ENCRYPTED_ENV, both, and neither under both MacOS and RedHat. A few tests were disabled/skipped for the MEM/ENCRYPTED cases. The error_handler_fs_test fails/hangs for MEM_ENV (presumably a timing problem) and I will introduce another PR/issue to track that problem. (I will also push a change to disable those tests soon). There is one more test in DBTest2 that also fails which I need to investigate or skip before this PR is merged.
Theoretically, this PR should also allow the test suite to run against an Env loaded from the registry, though I do not have one to try it with currently.
Finally, once this is accepted, it would be nice if there was a CircleCI job to run these tests on a checkin so this effort does not become stale. I do not know how to do that, so if someone could write that job, it would be appreciated :)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7566
Reviewed By: zhichao-cao
Differential Revision: D24408980
Pulled By: jay-zhuang
fbshipit-source-id: 911b1554a4d0da06fd51feca0c090a4abdcb4a5f
Summary:
Add a threshold timestamp, full_history_ts_low_ of type `std::string*` to
`CompactionIterator`, so that RocksDB can also perform garbage collection during
compaction.
* If full_history_ts_low_ is nullptr, then compaction iterator does not perform
GC, preserving all timestamp history for all keys. Compaction iterator will
treat user key with different timestamps as different user keys.
* If full_history_ts_low_ is not nullptr, then compaction iterator performs
GC. GC will look at keys older than `*full_history_ts_low_` and determine their
eligibility based on factors including snapshots.
Current rules of GC:
* If an internal key is in the same snapshot as a previous counterpart
with the same user key, and this key is eligible for GC, and the key is
not single-delete or merge operand, then this key can be dropped. Note
that the previous internal key cannot be a merge operand either.
* If a tombstone is the most recent one in the earliest snapshot and it
is eligible for GC, and keyNotExistsBeyondLevel() is true, then this
tombstone can be dropped.
* If a tombstone is the most recent one in a snapshot and it is eligible
for GC, and the compaction is at bottommost level, then all other older
internal keys of the same user key must also be eligible for GC, thus
can be dropped
* Single-delete, delete-range and merge are not currently supported.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7556
Test Plan: make check
Reviewed By: ltamasi
Differential Revision: D24507728
Pulled By: riversand963
fbshipit-source-id: 3c09c7301f41eed76dfcf4d1527e68cf6e0a8bb3
Summary:
This option determines whether WALs will be tracked in MANIFEST and verified on recovery.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7275
Test Plan:
db_options_test
options_test
Reviewed By: pdillinger
Differential Revision: D23181418
Pulled By: cheng-chang
fbshipit-source-id: 5dd1cdc166f3dfc1c93c094df4a2f7734e3b4547
Summary:
This PR merges the functionality of making the ColumnFamilyOptions, TableFactory, and DBOptions into Configurable into a single PR, resolving any merge conflicts
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5753
Reviewed By: ajkr
Differential Revision: D23385030
Pulled By: zhichao-cao
fbshipit-source-id: 8b977a7731556230b9b8c5a081b98e49ee4f160a
Summary:
Cleans up some of the dependencies on test code in the Makefile while building tools:
- Moves the test::RandomString, DBBaseTest::RandomString into Random
- Moves the test::RandomHumanReadableString into Random
- Moves the DestroyDir method into file_utils
- Moves the SetupSyncPointsToMockDirectIO into sync_point.
- Moves the FaultInjection Env and FS classes under env
These changes allow all of the tools to build without dependencies on test_util, thereby simplifying the build dependencies. By moving the FaultInjection code, the dependency in db_stress on different libraries for debug vs release was eliminated.
Tested both release and debug builds via Make and CMake for both static and shared libraries.
More work remains to clean up how the tools are built and remove some unnecessary dependencies. There is also more work that should be done to get the Makefile and CMake to align in their builds -- what is in the libraries and the sizes of the executables are different.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7097
Reviewed By: riversand963
Differential Revision: D22463160
Pulled By: pdillinger
fbshipit-source-id: e19462b53324ab3f0b7c72459dbc73165cc382b2
Summary:
Delicious copy-pasta from https://github.com/facebook/rocksdb/issues/7039
Also fixing DestroyDir to allow files to go missing while it is operating. This seems to fix failures I got with test plan reproducer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7103
Test Plan:
make blackbox_crash_test_with_atomic_flush for a while with
checkpoint_one_in=100
Reviewed By: siying
Differential Revision: D22435315
Pulled By: pdillinger
fbshipit-source-id: 0ec0538402493887aeda43ecc03f32979cb84ced
Summary:
The ```for``` loop in ```VerifyChecksumInBlocks``` only checks ```index_iter->Valid()``` which could be ```false``` either due to reaching the end of the index or, in case of partitioned index, it could be due to a checksum mismatch error when reading a 2nd level index block. Instead of throwing away the index iterator status, we need to return any errors back to the caller.
Tests:
Add a test in block_based_table_reader_test.cc.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6909
Reviewed By: pdillinger
Differential Revision: D21833922
Pulled By: anand1976
fbshipit-source-id: bc778ebf1121dbbdd768689de5183f07a9f0beae
Summary:
Currently, `DeleteDir` only deletes the directory if there are no other directories under the target dir. This PR makes it delete directories recursively.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6934
Test Plan:
Added a new unit test in testutil_test.cc.
`make testutil_test`
Reviewed By: zhichao-cao
Differential Revision: D21884211
Pulled By: cheng-chang
fbshipit-source-id: 0b9a48a200f494ee007aef5d1763b4aa331f8b5a
Summary:
Currently, in direct IO mode, `MultiGet` retrieves the data blocks one by one instead of in parallel, see `BlockBasedTable::RetrieveMultipleBlocks`.
Since direct IO is supported in `RandomAccessFileReader::MultiRead` in https://github.com/facebook/rocksdb/pull/6446, this PR applies `MultiRead` to `MultiGet` so that the data blocks can be retrieved in parallel.
Also, in direct IO mode and when data blocks are compressed and need to uncompressed, this PR only allocates one continuous aligned buffer to hold the data blocks, and then directly uncompress the blocks to insert into block cache, there is no longer intermediate copies to scratch buffers.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6815
Test Plan:
1. added a new unit test `BlockBasedTableReaderTest::MultiGet`.
2. existing unit tests and stress tests contain tests against `MultiGet` in direct IO mode.
Reviewed By: anand1976
Differential Revision: D21426347
Pulled By: cheng-chang
fbshipit-source-id: b8446ae0e74152444ef9111e97f8e402ac31b24f
Summary:
In crash test, the db directory might be set to /dev/shm or /tmp, in certain environments such as internal testing infrastructure, neither of these directories support direct IO, so direct IO is never enabled in crash test.
This PR sets up SyncPoints in direct IO related code paths to disable O_DIRECT flag in calls to `open`, so the direct IO code paths will be executed, all direct IO related assertions will be checked, but no real direct IO request will be issued to the file system.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6727
Test Plan:
export CRASH_TEST_EXT_ARGS="--use_direct_reads=1 --mmap_read=0"
make -j24 crash_test
Reviewed By: zhichao-cao
Differential Revision: D21139250
Pulled By: cheng-chang
fbshipit-source-id: db9adfe78d91aa4759835b1af91c5db7b27b62ee
Summary:
In https://github.com/facebook/rocksdb/pull/6455, we modified the interface of `RandomAccessFileReader::Read` to be able to get rid of memcpy in direct IO mode.
This PR applies the new interface to `BlockFetcher` when reading blocks from SST files in direct IO mode.
Without this PR, in direct IO mode, when fetching and uncompressing compressed blocks, `BlockFetcher` will first copy the raw compressed block into `BlockFetcher::compressed_buf_` or `BlockFetcher::stack_buf_` inside `RandomAccessFileReader::Read` depending on the block size. then during uncompressing, it will copy the uncompressed block into `BlockFetcher::heap_buf_`.
In this PR, we get rid of the first memcpy and directly uncompress the block from `direct_io_buf_` to `heap_buf_`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6689
Test Plan: A new unit test `block_fetcher_test` is added.
Reviewed By: anand1976
Differential Revision: D21006729
Pulled By: cheng-chang
fbshipit-source-id: 2370b92c24075692423b81277415feb2aed5d980
Summary:
Towards making compaction logic compatible with user timestamp.
When computing boundaries and overlapping ranges for inputs of compaction, We need to compare SSTs by user key without timestamp.
Test plan (devserver):
```
make check
```
Several individual tests:
```
./version_set_test --gtest_filter=VersionStorageInfoTimestampTest.GetOverlappingInputs
./db_with_timestamp_compaction_test
./db_with_timestamp_basic_test
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6645
Reviewed By: ltamasi
Differential Revision: D20960012
Pulled By: riversand963
fbshipit-source-id: ad377fa9eb481bf7a8a3e1824aaade48cdc653a4
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:
When paranoid_checks is on, DBImpl::CheckConsistency() iterates over all sst files and calls Env::GetFileSize() for each of them. As far as I could understand, this is pretty arbitrary and doesn't affect correctness - if filesystem doesn't corrupt fsynced files, the file sizes will always match; if it does, it may as well corrupt contents as well as sizes, and rocksdb doesn't check contents on open.
If there are thousands of sst files, getting all their sizes takes a while. If, on top of that, Env is overridden to use some remote storage instead of local filesystem, it can be *really* slow and overload the remote storage service. This PR adds an option to not do GetFileSize(); instead it does GetChildren() for parent directory to check that all the expected sst files are at least present, but doesn't check their sizes.
We can't just disable paranoid_checks instead because paranoid_checks do a few other important things: make the DB read-only on write errors, print error messages on read errors, etc.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6353
Test Plan: ran the added sanity check unit test. Will try it out in a LogDevice test cluster where the GetFileSize() calls are causing a lot of trouble.
Differential Revision: D19656425
Pulled By: al13n321
fbshipit-source-id: c2c421b367633033760d1f56747bad206d1fbf82
Summary:
unordered_write is incompatible with non-zero max_successive_merges. Although we check this at runtime, we currently don't prevent the user from setting this combination in options. This has led to stress tests to fail with this combination is tried in ::SetOptions.
The patch fixes that and also reverts the changes performed by https://github.com/facebook/rocksdb/pull/6254, in which max_successive_merges was mistakenly declared incompatible with unordered_write.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6284
Differential Revision: D19356115
Pulled By: maysamyabandeh
fbshipit-source-id: f06dadec777622bd75f267361c022735cf8cecb6
Summary:
allow_concurrent_memtable_write is incompatible with non-zero max_successive_merges. Although we check this at runtime, we currently don't prevent the user from setting this combination in options. This has led to stress tests to fail with this combination is tried in ::SetOptions. The patch fixes that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6254
Differential Revision: D19265819
Pulled By: maysamyabandeh
fbshipit-source-id: 47f2e2dc26fe0972c7152f4da15dadb9703f1179
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:
Adds an improved, replacement Bloom filter implementation (FastLocalBloom) for full and partitioned filters in the block-based table. This replacement is faster and more accurate, especially for high bits per key or millions of keys in a single filter.
Speed
The improved speed, at least on recent x86_64, comes from
* Using fastrange instead of modulo (%)
* Using our new hash function (XXH3 preview, added in a previous commit), which is much faster for large keys and only *slightly* slower on keys around 12 bytes if hashing the same size many thousands of times in a row.
* Optimizing the Bloom filter queries with AVX2 SIMD operations. (Added AVX2 to the USE_SSE=1 build.) Careful design was required to support (a) SIMD-optimized queries, (b) compatible non-SIMD code that's simple and efficient, (c) flexible choice of number of probes, and (d) essentially maximized accuracy for a cache-local Bloom filter. Probes are made eight at a time, so any number of probes up to 8 is the same speed, then up to 16, etc.
* Prefetching cache lines when building the filter. Although this optimization could be applied to the old structure as well, it seems to balance out the small added cost of accumulating 64 bit hashes for adding to the filter rather than 32 bit hashes.
Here's nominal speed data from filter_bench (200MB in filters, about 10k keys each, 10 bits filter data / key, 6 probes, avg key size 24 bytes, includes hashing time) on Skylake DE (relatively low clock speed):
$ ./filter_bench -quick -impl=2 -net_includes_hashing # New Bloom filter
Build avg ns/key: 47.7135
Mixed inside/outside queries...
Single filter net ns/op: 26.2825
Random filter net ns/op: 150.459
Average FP rate %: 0.954651
$ ./filter_bench -quick -impl=0 -net_includes_hashing # Old Bloom filter
Build avg ns/key: 47.2245
Mixed inside/outside queries...
Single filter net ns/op: 63.2978
Random filter net ns/op: 188.038
Average FP rate %: 1.13823
Similar build time but dramatically faster query times on hot data (63 ns to 26 ns), and somewhat faster on stale data (188 ns to 150 ns). Performance differences on batched and skewed query loads are between these extremes as expected.
The only other interesting thing about speed is "inside" (query key was added to filter) vs. "outside" (query key was not added to filter) query times. The non-SIMD implementations are substantially slower when most queries are "outside" vs. "inside". This goes against what one might expect or would have observed years ago, as "outside" queries only need about two probes on average, due to short-circuiting, while "inside" always have num_probes (say 6). The problem is probably the nastily unpredictable branch. The SIMD implementation has few branches (very predictable) and has pretty consistent running time regardless of query outcome.
Accuracy
The generally improved accuracy (re: Issue https://github.com/facebook/rocksdb/issues/5857) comes from a better design for probing indices
within a cache line (re: Issue https://github.com/facebook/rocksdb/issues/4120) and improved accuracy for millions of keys in a single filter from using a 64-bit hash function (XXH3p). Design details in code comments.
Accuracy data (generalizes, except old impl gets worse with millions of keys):
Memory bits per key: FP rate percent old impl -> FP rate percent new impl
6: 5.70953 -> 5.69888
8: 2.45766 -> 2.29709
10: 1.13977 -> 0.959254
12: 0.662498 -> 0.411593
16: 0.353023 -> 0.0873754
24: 0.261552 -> 0.0060971
50: 0.225453 -> ~0.00003 (less than 1 in a million queries are FP)
Fixes https://github.com/facebook/rocksdb/issues/5857
Fixes https://github.com/facebook/rocksdb/issues/4120
Unlike the old implementation, this implementation has a fixed cache line size (64 bytes). At 10 bits per key, the accuracy of this new implementation is very close to the old implementation with 128-byte cache line size. If there's sufficient demand, this implementation could be generalized.
Compatibility
Although old releases would see the new structure as corrupt filter data and read the table as if there's no filter, we've decided only to enable the new Bloom filter with new format_version=5. This provides a smooth path for automatic adoption over time, with an option for early opt-in.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6007
Test Plan: filter_bench has been used thoroughly to validate speed, accuracy, and correctness. Unit tests have been carefully updated to exercise new and old implementations, as well as the logic to select an implementation based on context (format_version).
Differential Revision: D18294749
Pulled By: pdillinger
fbshipit-source-id: d44c9db3696e4d0a17caaec47075b7755c262c5f
Summary:
file_reader_writer.h and .cc contain several files and helper function, and it's hard to navigate. Separate it to multiple files and put them under file/
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5803
Test Plan: Build whole project using make and cmake.
Differential Revision: D17374550
fbshipit-source-id: 10efca907721e7a78ed25bbf74dc5410dea05987
Summary:
MyRocks currently sets `max_write_buffer_number_to_maintain` in order to maintain enough history for transaction conflict checking. The effectiveness of this approach depends on the size of memtables. When memtables are small, it may not keep enough history; when memtables are large, this may consume too much memory.
We are proposing a new way to configure memtable list history: by limiting the memory usage of immutable memtables. The new option is `max_write_buffer_size_to_maintain` and it will take precedence over the old `max_write_buffer_number_to_maintain` if they are both set to non-zero values. The new option accounts for the total memory usage of flushed immutable memtables and mutable memtable. When the total usage exceeds the limit, RocksDB may start dropping immutable memtables (which is also called trimming history), starting from the oldest one.
The semantics of the old option actually works both as an upper bound and lower bound. History trimming will start if number of immutable memtables exceeds the limit, but it will never go below (limit-1) due to history trimming.
In order the mimic the behavior with the new option, history trimming will stop if dropping the next immutable memtable causes the total memory usage go below the size limit. For example, assuming the size limit is set to 64MB, and there are 3 immutable memtables with sizes of 20, 30, 30. Although the total memory usage is 80MB > 64MB, dropping the oldest memtable will reduce the memory usage to 60MB < 64MB, so in this case no memtable will be dropped.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5022
Differential Revision: D14394062
Pulled By: miasantreble
fbshipit-source-id: 60457a509c6af89d0993f988c9b5c2aa9e45f5c5
Summary:
Current PosixLogger performs IO operations using posix calls. Thus the
current implementation will not work for non-posix env. Created a new
logger class EnvLogger that uses env specific WritableFileWriter for IO operations.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5491
Test Plan: make check
Differential Revision: D15909002
Pulled By: ggaurav28
fbshipit-source-id: 13a8105176e8e42db0c59798d48cb6a0dbccc965
Summary:
The first key is used to defer reading the data block until this file gets to the top of merging iterator's heap. For short range scans, most files never make it to the top of the heap, so this change can reduce read amplification by a lot sometimes.
Consider the following workload. There are a few data streams (we'll be calling them "logs"), each stream consisting of a sequence of blobs (we'll be calling them "records"). Each record is identified by log ID and a sequence number within the log. RocksDB key is concatenation of log ID and sequence number (big endian). Reads are mostly relatively short range scans, each within a single log. Writes are mostly sequential for each log, but writes to different logs are randomly interleaved. Compactions are disabled; instead, when we accumulate a few tens of sst files, we create a new column family and start writing to it.
So, a typical sst file consists of a few ranges of blocks, each range corresponding to one log ID (we use FlushBlockPolicy to cut blocks at log boundaries). A typical read would go like this. First, iterator Seek() reads one block from each sst file. Then a series of Next()s move through one sst file (since writes to each log are mostly sequential) until the subiterator reaches the end of this log in this sst file; then Next() switches to the next sst file and reads sequentially from that, and so on. Often a range scan will only return records from a small number of blocks in small number of sst files; in this case, the cost of initial Seek() reading one block from each file may be bigger than the cost of reading the actually useful blocks.
Neither iterate_upper_bound nor bloom filters can prevent reading one block from each file in Seek(). But this PR can: if the index contains first key from each block, we don't have to read the block until this block actually makes it to the top of merging iterator's heap, so for short range scans we won't read any blocks from most of the sst files.
This PR does the deferred block loading inside value() call. This is not ideal: there's no good way to report an IO error from inside value(). As discussed with siying offline, it would probably be better to change InternalIterator's interface to explicitly fetch deferred value and get status. I'll do it in a separate PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5289
Differential Revision: D15256423
Pulled By: al13n321
fbshipit-source-id: 750e4c39ce88e8d41662f701cf6275d9388ba46a
Summary:
Currently we validate options in DB::Open. However the validation step is missing when options are dynamically updated in ::SetOptions. The patch fixes that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5368
Differential Revision: D15540101
Pulled By: maysamyabandeh
fbshipit-source-id: d27bbffd8f0252d1b50bcf59e0a70a278ed937f4
Summary:
There are too many types of files under util/. Some test related files don't belong to there or just are just loosely related. Mo
ve them to a new directory test_util/, so that util/ is cleaner.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5377
Differential Revision: D15551366
Pulled By: siying
fbshipit-source-id: 0f5c8653832354ef8caa31749c0143815d719e2c
