Summary:
CPU profiling reveals GetApproximateSizes as a bottleneck for performance. The current implementation is sub-optimal, it scans every file in every level to compute the result.
We can take advantage of the fact that all levels above 0 are sorted in the increasing order of key ranges and use binary search to locate the starting index. This can reduce the number of comparisons required to compute the result.
Test Plan: We have good test coverage. Run the tests.
Reviewers: sdong, igor, rven, dynamike
Subscribers: dynamike, maykov, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37755
Summary: Before the fix we also marked the bottommost level for compaction. This is wrong because then RocksDB has N+1 levels instead of N as before the compaction.
Test Plan: SuggestCompactRangeTest in db_test
Reviewers: yhchiang, rven
Reviewed By: rven
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37869
Summary: Remove duplicate code. If this diff looks good, I will cleanup other call sites as well.
Test Plan: unit tests
Reviewers: rven, yhchiang, igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37761
Summary:
Added these events:
* Recovery start, finish and also when recovery creates a file
* Trivial move
* Compaction start, finish and when compaction creates a file
* Flush start, finish
Also includes small fix to EventLogger
Also added option ROCKSDB_PRINT_EVENTS_TO_STDOUT which is useful when we debug things. I've spent far too much time chasing LOG files.
Still didn't get sst table properties in JSON. They are written very deeply into the stack. I'll address in separate diff.
TODO:
* Write specification. Let's first use this for a while and figure out what's good data to put here, too. After that we'll write spec
* Write tools that parse and analyze LOGs. This can be in python or go. Good intern task.
Test Plan: Ran db_bench with ROCKSDB_PRINT_EVENTS_TO_STDOUT. Here's the output: https://phabricator.fb.com/P19811976
Reviewers: sdong, yhchiang, rven, MarkCallaghan, kradhakrishnan, anthony
Reviewed By: anthony
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37521
Summary:
Make it build for CYGWIN.
Need to define "-std=gnu++11" instead of "-std=c++11" and use some replacement functions.
Test Plan: Build it and run some unit tests in CYGWIN
Reviewers: yhchiang, rven, anthony, kradhakrishnan, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37605
Summary:
CompactRange for universal compaction with num_levels > 1 seems to have a bug. The unit test also has a bug so it doesn't capture the problem.
Fix it. Revert the compact range to the logic equivalent to num_levels=1. Always compact all files together.
It should also fix DBTest.IncreaseUniversalCompactionNumLevels. The issue was that options.write_buffer_size = 100 << 10 and options.write_buffer_size = 100 << 10 are not used in later test scenarios. So write_buffer_size of 4MB was used. The compaction trigger condition is not anymore obvious as expected.
Test Plan: Run the new test and all test suites
Reviewers: yhchiang, rven, kradhakrishnan, anthony, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37551
Summary:
Based on feedback from D37083.
Are all of these correct? In some spaces it seems like we're doing SetMaxPossibleForUserKey() although we want the smallest possible internal key for user key.
Test Plan: make check
Reviewers: sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37341
Summary: Reading CompactionPicker I noticed this dangerous substraction of two unsigned integers. We should assert to mark this as safe.
Test Plan: make check
Reviewers: anthony, rven, yhchiang, sdong
Reviewed By: sdong
Subscribers: kradhakrishnan, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37041
Summary:
This diff implements a new `DB` method `PromoteL0` which moves all files in L0
to a given level skipping compaction, provided that the files have disjoint
ranges and all levels up to the target level are empty.
This method provides finer-grain control for trivial compactions, and it is
useful for bulk-loading pre-sorted keys. Compared to D34797, it does not change
the semantics of an existing operation, which can impact existing code.
PromoteL0 is designed to work well in combination with the proposed
`GetSstFileWriter`/`AddFile` interface, enabling to "design" the level structure
by populating one level at a time. Such fine-grained control can be very useful
for static or mostly-static databases.
Test Plan: `make check`
Reviewers: IslamAbdelRahman, philipp, MarkCallaghan, yhchiang, igor, sdong
Reviewed By: sdong
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37107
Summary: Add more logging to help debugging issues.
Test Plan: Run test suites
Reviewers: yhchiang, rven, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37401
Summary: To further distinguish the corruption cases were caused by storage media or in memory states when writing it, add a paranoid check after writing the file to iterate all the rows.
Test Plan: Add a new unit test for it
Reviewers: rven, igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37335
Summary:
A couple of times on Travis, we have had the thread status say that there were no compactions done and since we assert for it, the test failed.
We now fix this by waiting till compaction started.
Test Plan:
run DBTEST::*PreShutdown*
d=/tmp/j; rm -rf $d; seq 200 | parallel --gnu --eta 'd=/tmp/j/d-{}; mkdir -p $d; TEST_TMPDIR=$d ./db_test --gtest_filter=DBTest.PreShutdown* >& '$d'/log-{}'
Reviewers: sdong, igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37545
Summary:
Some Mongo+Rocks datasets in Parse's environment are not doing compactions very frequently. During the quiet period (with no IO), we'd like to schedule compactions so that our reads become faster. Also, aggressively compacting during quiet periods helps when write bursts happen. In addition, we also want to compact files that are containing deleted key ranges (like old oplog keys).
All of this is currently not possible with CompactRange() because it's single-threaded and blocks all other compactions from happening. Running CompactRange() risks an issue of blocking writes because we generate too much Level 0 files before the compaction is over. Stopping writes is very dangerous because they hold transaction locks. We tried running manual compaction once on Mongo+Rocks and everything fell apart.
MarkForCompaction() solves all of those problems. This is very light-weight manual compaction. It is lower priority than automatic compactions, which means it shouldn't interfere with background process keeping the LSM tree clean. However, if no automatic compactions need to be run (or we have extra background threads available), we will start compacting files that are marked for compaction.
Test Plan: added a new unit test
Reviewers: yhchiang, rven, MarkCallaghan, sdong
Reviewed By: sdong
Subscribers: yoshinorim, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37083
Summary:
The usage I'm fixing here caused trouble on Fedora 21 when
compiling with the current gcc version 4.9.2 20150212 (Red Hat 4.9.2-6) (GCC):
db/write_controller_test.cc: In member function ‘virtual void rocksdb::WriteControllerTest_SanityTest_Test::TestBody()’:
db/write_controller_test.cc:23:165: error: converting ‘false’ to pointer type for argument 1 of ‘char testing::internal::IsNullLiteralHelper(testing::internal::Secret*)’ [-Werror=conversion-null]
ASSERT_EQ(false, controller.IsStopped());
^
This change was induced mechanically via:
git grep -l -E 'ASSERT_EQ\(false'|xargs perl -pi -e 's/ASSERT_EQ\(false, /ASSERT_FALSE(/'
git grep -l -E 'ASSERT_EQ\(true'|xargs perl -pi -e 's/ASSERT_EQ\(true, /ASSERT_TRUE(/'
Except for the three in utilities/backupable/backupable_db_test.cc for which
I ended up reformatting (joining lines) in the result.
As for why this problem is exhibited with that version of gcc, and none
of the others I've used (from 4.8.1 through gcc-5.0.0 and newer), I suspect
it's a bug in F21's gcc that has been fixed in gcc-5.0.0.
Test Plan:
"make" now succeed on Fedora 21
Reviewers: ljin, rven, igor.sugak, yhchiang, sdong, igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37329
Summary: this is not used anywhere
Test Plan: compiles
Reviewers: yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37053
Summary: If ExpandWhileOverlapping() we don't clear inputs. That's a bug introduced by my recent patch https://reviews.facebook.net/D36687. However, we have no tests covering ExpandWhileOverlapping(). I created a task t6771252 to add ExpandWhileOverlapping() tests.
Test Plan: make check
Reviewers: sdong, rven, yhchiang
Reviewed By: yhchiang
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37077
Summary: D36669 introduces a bug that trivial moved data is not going to specific level but the next level, which will incorrectly be level 1 for level 0 compaciton if base level is not level 1. Fixing it by appreciating the output level
Test Plan: Run all tests
Reviewers: MarkCallaghan, rven, yhchiang, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37119
Summary:
Recent change of DBTest.DynamicLevelCompressionPerLevel2 has a bug that the second sync point is not enabled. Fix it. Also add an assert for that.
Also, flush compression is not tracked in the test. Add it.
Test Plan: Build everything
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37101
Summary: When commiting the sync point interface change, didn't resolve the new occurance of the old interface in rebase. Fix it.
Test Plan: Build and see it pass
Reviewers: igor, yhchiang, rven, anthony, kradhakrishnan
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37095
Summary:
Allow users to give a callback function with parameter using sync point, so more complicated verification can be done in tests.
Use it in DBTest.DynamicLevelCompressionPerLevel2 so that failures will be more easy to debug.
Test Plan: Run all tests. Run DBTest.DynamicLevelCompressionPerLevel2 with valgrind check.
Reviewers: rven, yhchiang, anthony, kradhakrishnan, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D36999
Summary: https://reviews.facebook.net/D36963 made the debug build much faster and that triggered failures of CompactFilesOnLevelCompaction test. 3 out of 4 last tests on Jenkins failed. I'm disabling this test temporarily, since we likely know the reason why it's failing and there's already work in progress to address it -- https://reviews.facebook.net/D36225
Test Plan: none
Reviewers: sdong, rven, yhchiang, meyering
Reviewed By: meyering
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36993
Summary: We should use mocked-out env for these tests to make it more realiable. Added benefit is that instead of actually sleeping for 3 seconds, we can instead pretend to sleep and just increase time counters.
Test Plan: for i in `seq 100`; do ./wal_manager_test --gtest_filter=WalManagerTest.WALArchivalTtl ;done
Reviewers: rven, meyering
Reviewed By: meyering
Subscribers: meyering, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36951
Summary:
The problem is that sometimes two memtables will be compacted together into a single file. In that case, our assertion
ASSERT_EQ(NumTableFilesAtLevel(0), 5);
fails because same amount of data is in 4 files instead of 5. We should wait for flush so that we prevent two memtables merging into a single file.
Test Plan: `for i in `seq 20`; do mrtest FIFOCompactionTest; done` -- fails at least once before. fails zero times after.
Reviewers: rven
Reviewed By: rven
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36939
Summary:
1. it doesn't work
2. we're not using it
In the future, if we need general benchmark framework, we should probably use https://github.com/google/benchmark
Test Plan: make all
Reviewers: yhchiang, rven, anthony, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36777
Summary:
The goal of this diff is to make Compaction class easier to use. This should also make new compaction algorithms easier to write (like CompactFiles from @yhchiang and dynamic leveled and multi-leveled universal from @sdong).
Here are couple of things demonstrating that Compaction class is hard to use:
1. we have two constructors of Compaction class
2. there's this thing called grandparents_, but it appears to only be setup for leveled compaction and not compactfiles
3. it's easy to introduce a subtle and dangerous bug like this: D36225
4. SetupBottomMostLevel() is hard to understand and it shouldn't be. See this comment: afbafeaeae/db/compaction.cc (L236-L241). It also made it harder for @yhchiang to write CompactFiles, as evidenced by this: afbafeaeae/db/compaction_picker.cc (L204-L210)
The problem is that we create Compaction object, which holds a lot of state, and then pass it around to some functions. After those functions are done mutating, then we call couple of functions on Compaction object, like SetupBottommostLevel() and MarkFilesBeingCompacted(). It is very hard to see what's happening with all that Compaction's state while it's travelling across different functions. If you're writing a new PickCompaction() function you need to try really hard to understand what are all the functions you need to run on Compaction object and what state you need to setup.
My proposed solution is to make important parts of Compaction immutable after construction. PickCompaction() should calculate compaction inputs and then pass them onto Compaction object once they are finalized. That makes it easy to create a new compaction -- just provide all the parameters to the constructor and you're done. No need to call confusing functions after you created your object.
This diff doesn't fully achieve that goal, but it comes pretty close. Here are some of the changes:
* have one Compaction constructor instead of two.
* inputs_ is constant after construction
* MarkFilesBeingCompacted() is now private to Compaction class and automatically called on construction/destruction.
* SetupBottommostLevel() is gone. Compaction figures it out on its own based on the input.
* CompactionPicker's functions are not passing around Compaction object anymore. They are only passing around the state that they need.
Test Plan:
make check
make asan_check
make valgrind_check
Reviewers: rven, anthony, sdong, yhchiang
Reviewed By: yhchiang
Subscribers: sdong, yhchiang, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36687
Summary: Need to remember to unref MemTableList->current() before deleting.
Test Plan: ran test with valgrind
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36855
Summary:
Test failing due to a missing directory caused by a simple bug (did not run into this on my dev box since the path already existed).
We should look into deleting test::TmpDir() before each test run.
Test Plan: ran test
Reviewers: igor, yhchiang, meyering, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36831
Summary:
Fixed xfunc related compile errors in ROCKSDB_LITE
Now make OPT=-DROCKSDB_LITE shared_lib -j32 would work
Test Plan:
make clean
make OPT=-DROCKSDB_LITE shared_lib -j32
make clean
make OPT=-DROCKSDB_LITE static_lib -j32
Reviewers: sdong, igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36825
Summary: Add tests for MemTableList
Test Plan: run test
Reviewers: yhchiang, kradhakrishnan, sdong, igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36735
Summary:
Fix a compile error in ROCKSDB_LITE in db/db_impl.cc
related to internal_stats.
Test Plan: make OPT=-DROCKSDB_LITE shared_lib
Reviewers: sdong, igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36819
Summary:
Fix a compilation error in ROCKSDB_LITE in db/internal_stats.h
Other compilation errors will be fixed in a separate diff.
Test Plan: make OPT=-DROCKSDB_LITE
Reviewers: sdong, igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36807
Summary:
Add a test case:
Write some keys without sync, flush, write other keys and do sync. Before flush finishes, host crashes and unsync data is dropped.
Tag the new test as disabled since it is not passing.
Test Plan: Run the test
Reviewers: MarkCallaghan, rven, anthony, igor, kradhakrishnan
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D36741
Summary:
This fixes two problems:
1) the env should not be created twice when use_existing_db is false
2) the env dtor should run before cachedev_fd_ is closed.
Task ID: #
Blame Rev:
Test Plan:
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D36795
Summary: Other than making some class members private, this is a documentation-only change
Test Plan: unit tests
Reviewers: sdong, yhchiang, igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36567
Summary: Now EnvOptions uses unsanitized DB options. bytes_per_sync is tuned off when rate_limiter is used, but this change doesn't take effort.
Test Plan: See different I/O pattern in db_bench running fillseq.
Reviewers: yhchiang, kradhakrishnan, rven, anthony, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D36723
Summary: Now trivial move is only triggered when moving from level n to n+1. With dynamic level base, it is possible that file is moved from level 0 to level n, while levels from 1 to n-1 are empty. Extend trivial move to this case.
Test Plan: Add a more unit test of sequential loading. Non-trivial compaction happened without the patch and now doesn't happen.
Reviewers: rven, yhchiang, MarkCallaghan, igor
Reviewed By: igor
Subscribers: leveldb, dhruba, IslamAbdelRahman
Differential Revision: https://reviews.facebook.net/D36669
Summary:
There are some cases when flachcache file descriptor was
already allocated (i.e. fb-MySQL). Then NewFlashcacheAwareEnv returns an
error at open() because fd was already assigned. This diff adds another
function to instantiate FlashcacheAwareEnv, with pre-allocated fd cachedev_fd.
Test Plan: Tested with MyRocks using this function, then worked
Reviewers: sdong, igor
Reviewed By: igor
Subscribers: dhruba, MarkCallaghan, rven
Differential Revision: https://reviews.facebook.net/D36447
Summary: Now we add warnings when user configures compression and the compression is not supported.
Test Plan:
Configured compression to non-supported values. Observed messages in my log:
2015/03/26-12:17:57.586341 7ffb8a496840 [WARN] Compression type chosen for level 2 is not supported: LZ4. RocksDB will not compress data on level 2.
2015/03/26-12:19:10.768045 7f36f15c5840 [WARN] Compression type chosen is not supported: LZ4. RocksDB will not compress data.
Reviewers: rven, sdong, yhchiang
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D35979
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
Summary:
If accumulated_num_non_deletions_ were ever smaller than
accumulated_num_deletions_, the computation of
"accumulated_num_non_deletions_ - accumulated_num_deletions_"
would result in a logically "negative" value, but since
the two operands are unsigned (uint64_t), the result corresponding
to e.g., -1 would 2^64-1.
Instead, return 0 in that case.
Test Plan:
- ensure "make check" still passes
- temporarily add an "abort();" call in the new "if"-block, and
observe that it fails in some test cases. However, note that
this case is triggered only when the two numbers are equal.
Thus, no test case triggers the erroneous behavior this
change is designed to avoid. If anyone can construct a
scenario in which that bug would be triggered, I'll be
happy to add a test case.
Reviewers: ljin, igor, rven, igor.sugak, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D36489
Summary: Int is used for level size targets when options_.level_compaction_dynamic_level_bytes=true, which will cause overflow when database grows big. Fix it.
Test Plan: Add a new unit test which fails without the fix.
Reviewers: rven, yhchiang, MarkCallaghan, igor
Reviewed By: igor
Subscribers: leveldb, dhruba, yoshinorim
Differential Revision: https://reviews.facebook.net/D36453
Summary: In some db_test tests sync points are not cleared which will cause unexpected results in the next tests. Clean them up in test cleaning up.
Test Plan:
Run the same tests that used to fail:
build using USE_CLANG=1 and run
./db_test --gtest_filter="DBTest.CompressLevelCompaction:*DBTestUniversalCompactionParallel*"
Reviewers: rven, yhchiang, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D36429
krad