Summary:
Persistent cache tier is the tier abstraction that can work for any block
device based device mounted on a file system. The design/implementation can
handle any generic block device.
Any generic block support is achieved by generalizing the access patten as
{io-size, q-depth, direct-io/buffered}.
We have specifically tested and adapted the IO path for NVM and SSD.
Persistent cache tier consists of there parts :
1) File layout
Provides the implementation for handling IO path for reading and writing data
(key/value pair).
2) Meta-data
Provides the implementation for handling the index for persistent read cache.
3) Implementation
It binds (1) and (2) and flushed out the PersistentCacheTier interface
This patch provides implementation for (1)(2). Follow up patch will provide (3)
and tests.
Test Plan: Compile and run check
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D57117
* Added new statistics and refactored to allow ioptions to be passed around as required to access environment and statistics pointers (and, as a convenient side effect, info_log pointer).
* Prevent incrementing compression counter when compression is turned off in options.
* Prevent incrementing compression counter when compression is turned off in options.
* Added two more supported compression types to test code in db_test.cc
* Prevent incrementing compression counter when compression is turned off in options.
* Added new StatsLevel that excludes compression timing.
* Fixed casting error in coding.h
* Fixed CompressionStatsTest for new StatsLevel.
* Removed unused variable that was breaking the Linux build
Summary: Refactor cache.cc so that I can plugin clock cache (D55581). Mainly move `ShardedCache` to separate file, move `LRUHandle` back to cache.cc and rename it lru_cache.cc.
Test Plan:
make check -j64
Reviewers: lightmark, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D59655
Summary: In Zlib_Compress and BZip2_Compress the calculation for size was slightly off when using compression_foramt_version 2 (which includes the decompressed size in the output). Also there were unnecessary loops around the deflate/BZ2_bzCompress calls. In Zlib_Compress there was also a possible exit from the function after calling deflateInit2 that didn't call deflateEnd.
Test Plan: Standard tests
Reviewers: sdong, IslamAbdelRahman, igor
Reviewed By: igor
Subscribers: sdong, IslamAbdelRahman, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D60537
Summary:
I was investigating performance issues in the SstFileWriter and found all of the following:
- The SstFileWriter::Add() function created a local InternalKey every time it was called generating a allocation and free each time. Changed to have an InternalKey member variable that can be reset with the new InternalKey::Set() function.
- In SstFileWriter::Add() the smallest_key and largest_key values were assigned the result of a ToString() call, but it is simpler to just assign them directly from the user's key.
- The Slice class had no move constructor so each time one was returned from a function a new one had to be allocated, the old data copied to the new, and the old one was freed. I added the move constructor which also required a copy constructor and assignment operator.
- The BlockBuilder::CurrentSizeEstimate() function calculates the current estimate size, but was being called 2 or 3 times for each key added. I changed the class to maintain a running estimate (equal to the original calculation) so that the function can return an already calculated value.
- The code in BlockBuilder::Add() that calculated the shared bytes between the last key and the new key duplicated what Slice::difference_offset does, so I replaced it with the standard function.
- BlockBuilder::Add() had code to copy just the changed portion into the last key value (and asserted that it now matched the new key). It is more efficient just to copy the whole new key over.
- Moved this same code up into the 'if (use_delta_encoding_)' since the last key value is only needed when delta encoding is on.
- FlushBlockBySizePolicy::BlockAlmostFull calculated a standard deviation value each time it was called, but this information would only change if block_size of block_size_deviation changed, so I created a member variable to hold the value to avoid the calculation each time.
- Each PutVarint??() function has a buffer and calls std::string::append(). Two or three calls in a row could share a buffer and a single call to std::string::append().
Some of these will be helpful outside of the SstFileWriter. I'm not 100% the addition of the move constructor is appropriate as I wonder why this wasn't done before - maybe because of compiler compatibility? I tried it on gcc 4.8 and 4.9.
Test Plan: The changes should not affect the results so the existing tests should all still work and no new tests were added. The value of the changes was seen by manually testing the SstFileWriter class through MyRocks and adding timing code to identify problem areas.
Reviewers: sdong, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59607
Summary: this test support CLI option to select HdfsEnv/NativeHdfsEnv now. The latter one is default. add test about when Rename(src, target) should overwrite target
Test Plan: existing test
Reviewers: sdong, andrewkr
Reviewed By: andrewkr
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D60399
Summary:
Use @omegaga's awesome feature to avoid use of callbacks for ensuring
SyncPoints happen in a particular thread.
Depends on D60375.
Test Plan:
$ ./auto_roll_logger_test
Reviewers: omegaga, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, omegaga, leveldb
Differential Revision: https://reviews.facebook.net/D60471
Summary: Add markers to sync points. A marked sync point will only be active when it is on the same thread as the marker sync point.
Test Plan: Write a unit test to validate.
Reviewers: sdong, IslamAbdelRahman, andrewkr
Reviewed By: andrewkr
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D60375
Summary: Add option write_buffer_manager to help users control total memory spent on memtables across multiple DB instances.
Test Plan: Add a new unit test.
Reviewers: yhchiang, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: adela, benj, sumeet, muthu, leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59925
Summary: LockFile is unnecessary in unit test
Test Plan: env_basic_test.cc
Reviewers: andrewkr
Reviewed By: andrewkr
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D60285
Summary:
Previously we couldn't run env_basic_test on Env::Default (PosixEnv on
our platforms) since GetChildren*() behavior was inconsistent with our other
Envs. We can normalize the output of GetChildren*() such that these test cases
work on PosixEnv too.
Test Plan: ran env_basic_test
Reviewers: wanning
Reviewed By: wanning
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D59943
Summary:
move cleanup to TearDown and handle directories, so cleanup will happen
even if a test fails in the middle.
Test Plan: ./env_basic_test
Reviewers: wanning
Reviewed By: wanning
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D60243
Summary: ensure no 2nd level children under test_dir_
Test Plan: env_basic_test on 4 envs
Reviewers: andrewkr
Reviewed By: andrewkr
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59979
Summary:
Add a read option `background_purge_on_iterator_cleanup` to avoid deleting files in foreground when destroying iterators.
Instead, a job is scheduled in high priority queue and would be executed in a separate background thread.
Test Plan: Add a variant of PurgeObsoleteFileTest. Turn on background purge option in the new test, and use sleeping task to ensure files are deleted in background.
Reviewers: IslamAbdelRahman, sdong
Reviewed By: IslamAbdelRahman
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59499
Summary: filter_deltes is not a frequently used feature. Remove it.
Test Plan: Run all test suites.
Reviewers: igor, yhchiang, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59427
Summary: We introduced default slow down and stop condition, but didn't reset it in bulk load mode. Fix it.
Test Plan: N/A
Reviewers: igor, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59757
Summary: Fixed a crash bug that incorrectly parse deprecated options in options_helper
Test Plan:
run db_bench with an old options file with memtable_prefix_bloom_probes
./db_bench --options_file=AN_OLD_OPTIONS_FILE --num=100 --benchmarks=fillseq
Reviewers: sdong, IslamAbdelRahman
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D59787
Summary:
Add option to not flush memtable on open()
In case the option is enabled, don't delete existing log files by not updating log numbers to MANIFEST.
Will still flush if we need to (e.g. memtable full in the middle). In that case we also flush final memtable.
If wal_recovery_mode = kPointInTimeRecovery, do not halt immediately after encounter corruption. Instead, check if seq id of next log file is last_log_sequence + 1. In that case we continue recovery.
Test Plan: See unit test.
Reviewers: dhruba, horuff, sdong
Reviewed By: sdong
Subscribers: benj, yhchiang, andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D57813
Summary:
Try to decompress compressed blocks when a special flag is set.
assert and crash in debug builds if we can't decompress the just-compressed input.
Test Plan: Run unit-tests.
Reviewers: dhruba, andrewkr, sdong, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59145
Summary: It confuses some compilers to have slice.cc under multiple directories. Merge them.
Test Plan: Run existing tests
Reviewers: andrewkr, yhchiang, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59409
Summary: delete_scheduler_test and db_sst_test share a same directory name, causing possible fails on both tests when running in parallel. Fixed by changing directory name.
Test Plan: Run the two tests in parallel: `parallel -u ./{} ::: delete_scheduler_test db_sst_test`
Reviewers: sdong, andrewkr
Reviewed By: sdong, andrewkr
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59529
Summary:
memtable_prefix_bloom_probes is not a critical option. Remove it to reduce number of options.
It's easier for users to make mistakes with memtable_prefix_bloom_bits, turn it to memtable_prefix_bloom_bits_ratio
Test Plan: Run all existing tests
Reviewers: yhchiang, igor, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: gunnarku, yoshinorim, MarkCallaghan, leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59199
Summary: AFIK, options builder is not used by anyone. Remove it.
Test Plan: Run all existing tests.
Reviewers: IslamAbdelRahman, andrewkr, igor
Reviewed By: igor
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59319
GCC 5.4 will complain (see also options_parser.cc):
/home/abuild/rpmbuild/BUILD/arangodb-3.0.0r1/3rdParty/rocksdb/rocksdb/util/options_builder.cc: In function 'rocksdb::CompactionStyle rocksdb::{anonymous}::PickCompactionStyle(size_t, int, int, uint64_t)':
/home/abuild/rpmbuild/BUILD/arangodb-3.0.0r1/3rdParty/rocksdb/rocksdb/util/options_builder.cc:29:7: error: 'log' is not a member of 'std'
std::log(target_db_size / write_buffer_size) / std::log(kBytesForLevelMultiplier)));
^
/home/abuild/rpmbuild/BUILD/arangodb-3.0.0r1/3rdParty/rocksdb/rocksdb/util/options_builder.cc:29:7: note: suggested alternative:
In file included from /usr/include/features.h:365:0,
from /usr/include/math.h:26,
from /home/abuild/rpmbuild/BUILD/arangodb-3.0.0r1/3rdParty/rocksdb/rocksdb/util/options_builder.cc:6:
/usr/include/bits/mathcalls.h:109:1: note: 'log'
__MATHCALL_VEC (log,, (_Mdouble_ __x));
Summary:
- Provide env_test as a static library. We will build it for future releases so internal Envs can use env_test by linking against this library.
- Add tests for CustomEnv, which is configurable via ENV_TEST_URI environment variable. It uses the URI-based Env lookup (depends on D58449).
- Refactor env_basic_test cases to use a unique/configurable directory for test files.
Test Plan:
built a test binary against librocksdb_env_test.a. It registered the
default Env with URI prefix "a://".
- verify runs all CustomEnv tests when URI with correct prefix is provided
```
$ ENV_TEST_URI="a://ok" ./tmp --gtest_filter="CustomEnv/*"
...
[ PASSED ] 12 tests.
```
- verify runs no CustomEnv tests when URI with non-matching prefix is provided
```
$ ENV_TEST_URI="b://ok" ./tmp --gtest_filter="CustomEnv/*"
...
[ PASSED ] 0 tests.
```
Reviewers: ldemailly, IslamAbdelRahman, lightmark, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D58485
Summary:
Extracted basic Env-related tests from mock_env_test and memenv_test into a
parameterized test for Envs: env_basic_test.
Depends on D58449. (The dependency is here only so I can keep this series of
diffs in a chain -- there is no dependency on that diff's code.)
Test Plan: ran tests
Reviewers: IslamAbdelRahman, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D58635
Summary:
Direct IO checkin breaks Windows build. Fixing the code to work for
Windows.
Test Plan: Run env_test in Windows 10 and make check in Linux
Reviewers: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D59073
Summary: When rate_bytes_per_sec * refill_period_us_ overflows, the actual limited rate is very low. Handle this case so the rate will be large.
Test Plan: Add a unit test for it.
Reviewers: IslamAbdelRahman, andrewkr
Reviewed By: andrewkr
Subscribers: yiwu, lightmark, leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D58929
Summary:
Made it consistent with posix Env, which uses pread() that returns 0
(success) when an offset is given beyond EOF. The purpose of making these Envs
behave consistently is I am repurposing the in-memory Envs' tests for the basic
Env tests in D58635.
Test Plan: ran mock_env_test and memenv_test
Reviewers: IslamAbdelRahman, lightmark, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D58845
Summary: util/threadpool.cc's function name is the same as a well-known class name. It breaks unity build. Rename it.
Test Plan: Run all existing test.
Reviewers: yiwu, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D58881
Summary:
O_DIRECT is not available in Mac as a flag for open. The fix is to make
use of fctl after the file is opened
Test Plan: Run the tests on mac and Linux
Reviewers: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D58665
Summary: Make CompactionOptionsFIFO a part of mutable_cf_options
Test Plan: UT
Reviewers: sdong
Reviewed By: sdong
Subscribers: andrewkr, lgalanis, dhruba
Differential Revision: https://reviews.facebook.net/D58653
Summary:
This patch adds direct IO capability to RocksDB Env.
The direct IO capability is required for persistent cache since NVM is best
accessed as 4K direct IO. SSDs can leverage direct IO for reading.
Direct IO requires the offset and size be sector size aligned, and memory to
be kernel page aligned. Since neither RocksDB/Persistent read cache data
layout is aligned to sector size, the code can accommodate reading unaligned IO size
(or unaligned memory) at the cost of an alloc/copy.
The write code path expects the size and memory to be aligned.
Test Plan: Run RocksDB unit tests
Reviewers: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D57393
Summary:
Persistent read cache relies on the accuracy of the GetUniqueIdFromFile
to generate a unique key for a given block of data. Currently we don't have an
implementation for Mac.
This patch adds an implementation.
Test Plan: Run unit tests
Reviewers: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D58413
Summary:
Expose a simple function to convert CompressionType to it's corresponding option string
This is for a diff @yoshinorim is working on for MyRocks
Test Plan: unittest
Reviewers: yhchiang, andrewkr, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, yoshinorim
Differential Revision: https://reviews.facebook.net/D58215
Summary:
Added a new abstraction to cache page to RocksDB designed for the read
cache use.
RocksDB current block cache is more of an object cache. For the persistent read cache
project, what we need is a page cache equivalent. This changes adds a cache
abstraction to RocksDB to cache pages called PersistentCache. PersistentCache can cache
uncompressed pages or raw pages (content as in filesystem). The user can
choose to operate PersistentCache either in COMPRESSED or UNCOMPRESSED mode.
Blame Rev:
Test Plan: Run unit tests
Reviewers: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D55707
Conditionally retrofit thread_posix for use with std::thread
and reuse the same logic. Posix users continue using Posix interfaces.
Enable XPRESS compression in test runs.
Fix master introduced signed/unsigned mismatch.
Summary:
Consider the following WAL with 4 batch entries prefixed with their sequence at time of memtable insert.
[1: BEGIN_PREPARE, PUT, PUT, PUT, PUT, END_PREPARE(a)]
[1: BEGIN_PREPARE, PUT, PUT, PUT, PUT, END_PREPARE(b)]
[4: COMMIT(a)]
[7: COMMIT(b)]
The first two batches do not consume any sequence numbers so are both prefixed with seq=1.
For 2pc commit, memtable insertion takes place before COMMIT batch is written to WAL.
We can see that sequence number consumption takes place between WAL entries giving us the seemingly sparse sequence prefix for WAL entries.
This is a valid WAL.
Because with 2PC markers one WriteBatch points to another batch containing its inserts a writebatch can consume more or less sequence numbers than the number of sequence consuming entries that it contains.
We can see that, given the entries in the WAL, 6 sequence ids were consumed. Yet on recovery the maximum sequence consumed would be 7 + 3 (the number of sequence numbers consumed by COMMIT(b))
So, now upon recovery we must track the actual consumption of sequence numbers.
In the provided scenario there will be no sequence gaps, but it is possible to produce a sequence gap. This should not be a problem though. correct?
Test Plan: provided test.
Reviewers: sdong
Subscribers: andrewkr, leveldb, dhruba, hermanlee4
Differential Revision: https://reviews.facebook.net/D57645
krad