Commit Graph

325 Commits

Author SHA1 Message Date
Siying Dong
f3ae3d07cc Add more black-box tests for PlainTable and explicitly support total order mode
Summary:
1. Add some more implementation-aware tests for PlainTable
2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0  means binary search only. Also fixes some divide 0 risks.
3. Explicitly support total order (only use binary search)
4. some code cleaning up.

Test Plan: make all check

Reviewers: haobo, kailiu

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16023
2014-02-12 17:37:22 -08:00
kailiu
e6b3e3b4db Support prefix seek in UserCollectedProperties
Summary: We'll need the prefix seek support for property aggregation.

Test Plan: make all check

Reviewers: haobo, sdong, dhruba

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15963
2014-02-12 13:14:59 -08:00
Igor Canadi
ca5f1a225a CompactionContext to include is_manual_compaction
Summary: Added a bit more information to compaction context, requested by internal team at FB.

Test Plan: Modified CompactionFilter test to make sure is_manual_compaction is properly set.

Reviewers: haobo

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16095
2014-02-12 12:24:18 -08:00
Lei Jin
994c327b86 IOError cleanup
Summary: Clean up IOErrors so that it only indicates errors talking to device.

Test Plan: make all check

Reviewers: igor, haobo, dhruba, emayanke

Reviewed By: igor

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15831
2014-02-12 11:42:54 -08:00
Siying Dong
33042669f6 Reduce malloc of iterators in Get() code paths
Summary:
This patch optimized Get() code paths by avoiding malloc of iterators. Iterator creation is moved to mem table rep implementations, where a callback is called when any key is found. This is the same practice as what we do in (SST) table readers.

db_bench result for readrandom following a writeseq, with no compression, single thread and tmpfs, we see throughput improved to 144958 from 139027, about 3%.

Test Plan: make all check

Reviewers: dhruba, haobo, igor

Reviewed By: haobo

CC: leveldb, yhchiang

Differential Revision: https://reviews.facebook.net/D14685
2014-02-11 10:32:51 -08:00
Igor Canadi
8e634d3ea4 Merge pull request #74 from alberts/lz4
Support for LZ4 compression.
2014-02-10 15:46:56 -08:00
Igor Canadi
bc2ff597b8 Fixed wrong comment GetTableMetaData -> GetLiveFilesMetaData 2014-02-10 10:55:10 -08:00
Albert Strasheim
df2f92214a Support for LZ4 compression. 2014-02-08 14:15:51 -08:00
Dhruba Borthakur
0982c38020 Fix compilation error with gcc 4.7
Summary:
Fix compilation error with gcc 4.7

Test Plan:
make clean
make

Reviewers:

CC:

Task ID: #

Blame Rev:
2014-02-07 13:52:54 -08:00
Yueh-Hsuan Chiang
3ce8d9a988 Add support for plain table format to sst_dump.
Summary:
This diff enables the command line tool `sst_dump` to work for sst files
under plain table format.  Changes include:
  * In tools/sst_dump.cc:
    - add support for plain table format
    - display prefix_extractor information when --show_properties is on
  * In table/format.cc
    - Now the table magic number of a Footer can be later initialized
      via ReadFooterFromFile().
  * In table/meta_bocks:
    - add function ReadTableMagicNumber() that reads the magic number of
      the specified file.

Minor fixes:
 - remove a duplicate #include in table/table_test.cc
 - fix a commentary typo in include/rocksdb/memtablerep.h
 - fix lint errors.

Test Plan:
Runs sst_dump with both block-based and plain-table format files with
different arguments, specifically those with --show-properties and --from.

* sample output:
  https://reviews.facebook.net/P261

Reviewers: kailiu, sdong, xjin

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15903
2014-02-07 11:15:00 -08:00
kailiu
84f8185fc0 Merge branch 'master' into performance
Conflicts:
	HISTORY.md
	db/db_impl.cc
	db/memtable.cc
2014-02-05 21:21:00 -08:00
kailiu
d43ebd8c65 Put table factory back to public api
Summary:
Previous I am too ambitious to hide every detail about table factory
to internal api. However, we cannot pass the compilatoin for external
users since we use table factory as the shared_ptr, which requires
the definition of table factory's destructor.

Test Plan: make check;

Reviewers: sdong, haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15861
2014-02-03 19:51:20 -08:00
Igor Canadi
2966d764cd Fix some 32-bit compile errors
Summary: RocksDB doesn't compile on 32-bit architecture apparently. This is attempt to fix some of 32-bit errors. They are reported here: https://gist.github.com/paxos/8789697

Test Plan: RocksDB still compiles on 64-bit :)

Reviewers: kailiu

Reviewed By: kailiu

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15825
2014-02-03 13:48:30 -08:00
Siying Dong
d169b67680 [Performance Branch] PlainTable to encode rows with seqID 0, value type using 1 internal byte.
Summary: In PlainTable, use one single byte to represent 8 bytes of internal bytes, if seqID = 0 and it is value type (which should be common for bottom most files). It is to save 7 bytes for uncompressed cases.

Test Plan: make all check

Reviewers: haobo, dhruba, kailiu

Reviewed By: haobo

CC: igor, leveldb

Differential Revision: https://reviews.facebook.net/D15489
2014-02-03 12:19:30 -08:00
kailiu
4f6cb17bdb First phase API clean up
Summary:
Addressed all the issues in https://reviews.facebook.net/D15447.
Now most table-related modules are hidden from user land.

Test Plan: make check

Reviewers: sdong, haobo, dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15525
2014-02-03 00:30:43 -08:00
kailiu
4e0298f23c Clean up arena API
Summary:
Easy thing goes first. This patch moves arena to internal dir; based
on which, the coming patch will deal with memtable_rep.

Test Plan: make check

Reviewers: haobo, sdong, dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15615
2014-01-30 22:10:10 -08:00
Dhruba Borthakur
abd70ecc2b The default settings enable checksum verification on every read.
Summary: The default settings enable checksum verification on every read.

Test Plan: make check

Reviewers: haobo

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15591
2014-01-30 19:14:03 -08:00
Igor Canadi
ac92420fc5 Merge branch 'master' into performance
Conflicts:
	db/db_impl.h
2014-01-30 10:09:23 -08:00
kailiu
3170abd297 Remove unused classes
Summary: This is a followup diff for https://reviews.facebook.net/D15447, which picks the most simple task: delete some unused memtable reps.

Test Plan: make

Reviewers: haobo, dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15585
2014-01-29 16:40:36 -08:00
Igor Canadi
e5ec7384a0 Better interface to create BackupEngine
Summary: I think it looks nicer. In RocksDB we have both styles, but I think that static method is the more common version.

Test Plan: backupable_db_test

Reviewers: ljin, benj, swk

Reviewed By: ljin

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15519
2014-01-28 16:01:53 -08:00
Igor Canadi
ec2fa4a690 Export BackupEngine
Summary:
Lots of clients have problems with using StackableDB interface. It's nice to have BackupableDB as a layer on top of DB, but not necessary.

This diff exports BackupEngine, which can be used to create backups without forcing clients to use StackableDB interface.

Test Plan: backupable_db_test

Reviewers: dhruba, ljin, swk

Reviewed By: ljin

CC: leveldb, benj

Differential Revision: https://reviews.facebook.net/D15477
2014-01-28 11:26:07 -08:00
kailiu
a5e220f5ef Merge branch 'master' into performance
Conflicts:
	Makefile
	db/db_impl.cc
	db/db_test.cc
	db/memtable_list.cc
	db/memtable_list.h
	table/block_based_table_reader.cc
	table/table_test.cc
	util/cache.cc
	util/coding.cc
2014-01-28 10:35:55 -08:00
Igor Canadi
832158e7f7 Fsync directory after we create a new file
Summary:
@dhruba, I'm not sure where we need to sync the directory. I implemented the function in Env() and added the dir sync just after we close the newly created file in the builder.

Should I also add FsyncDir() to new files that get created by a compaction?

Test Plan: Confirmed that FsyncDir is returning Status::OK()

Reviewers: dhruba, haobo

Reviewed By: dhruba

CC: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D14751
2014-01-27 11:02:21 -08:00
Siying Dong
b20486f294 [Performance Branch] HashLinkList to avoid to convert length prefixed string back to internal keys
Summary: Converting from length prefixed buffer back to internal key costs some CPU but it is not necessary. In this patch, internal keys are pass though the functions so that we don't need to convert back to it.

Test Plan: make all check

Reviewers: haobo, kailiu

Reviewed By: kailiu

CC: igor, leveldb

Differential Revision: https://reviews.facebook.net/D15393
2014-01-27 10:26:14 -08:00
Siying Dong
8477255da3 Moving Some includes from options.h to forward declaration
Summary: By removing some includes form options.h and reply on forward declaration, we can more easily reason the dependencies.

Test Plan: make all check

Reviewers: kailiu, haobo, igor, dhruba

Reviewed By: kailiu

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15411
2014-01-24 17:16:22 -08:00
Igor Canadi
b13bdfa500 Add a call DisownData() to Cache, which should speed up shutdown
Summary: On a shutdown, freeing memory takes a long time. If we're shutting down, we don't really care about memory leaks. I added a call to Cache that will avoid freeing all objects in cache.

Test Plan:
I created a script to test the speedup and demonstrate how to use the call: https://phabricator.fb.com/P3864368

Clean shutdown took 7.2 seconds, while fast and dirty one took 6.3 seconds. Unfortunately, the speedup is not that big, but should be bigger with bigger block_cache. I have set up the capacity to 80GB, but the script filled up only ~7GB.

Reviewers: dhruba, haobo, MarkCallaghan, xjin

Reviewed By: dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15069
2014-01-24 14:57:52 -08:00
Lei Jin
aba2acb5ec CompactRange() to return status
Summary: as title

Test Plan:
make all check
What else tests shall I cover?

Reviewers: igor, haobo

CC:

Differential Revision: https://reviews.facebook.net/D15339
2014-01-23 16:41:46 -08:00
Kai Liu
054c5dda8c Merge branch 'master' into performance
Conflicts:
	db/db_impl.cc
	db/db_test.cc
	db/memtable.cc
	db/version_set.cc
	include/rocksdb/statistics.h
	util/statistics_imp.h
2014-01-23 16:32:49 -08:00
Tomislav Novak
81c9cc9b3b Tailing iterator
Summary:
This diff implements a special type of iterator that doesn't create a snapshot
(can be used to read newly inserted data) and is optimized for doing sequential
reads.

TailingIterator uses current superversion number to determine whether to
invalidate its internal iterators. If the version hasn't changed, it can often
avoid doing expensive seeks over immutable structures (sst files and immutable
memtables).

Test Plan:
* new unit tests
* running LD with this patch

Reviewers: igor, dhruba, haobo, sdong, kailiu

Reviewed By: sdong

CC: leveldb, lovro, march

Differential Revision: https://reviews.facebook.net/D15285
2014-01-23 16:26:08 -08:00
Kai Liu
bb19b530ca Aggressively inlining the short functions in coding.cc
Summary:
This diff takes an even more aggressive way to inline the functions. A decent rule that I followed is "not inline a function if it is more than 10 lines long."

Normally optimizing code by inline is ugly and hard to control, but since one of our usecase has significant amount of CPU used in functions from coding.cc, I'd like to try this diff out.

Test Plan:
1. the size for some .o file increased a little bit, but most less than 1%. So I think the negative impact of inline is negligible.
2. As the regression test shows (ran for 10 times and I calculated the average number)

    Metrics                                         Befor    After
    ========================================================================
    rocksdb.build.fillseq.qps                       426595   444515    (+4.6%)
    rocksdb.build.memtablefillrandom.qps            121739   123110
    rocksdb.build.memtablereadrandom.qps            1285103  1280520
    rocksdb.build.overwrite.qps                     125816   135570    (+9%)
    rocksdb.build.readrandom_fillunique_random.qps  285995   296863
    rocksdb.build.readrandom_memtable_sst.qps       1027132  1027279
    rocksdb.build.readrandom.qps                    1041427  1054665
    rocksdb.build.readrandom_smallblockcache.qps    1028631  1038433
    rocksdb.build.readwhilewriting.qps              918352   914629

Reviewers: haobo, sdong, igor

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15291
2014-01-23 16:03:34 -08:00
Igor Canadi
83681bf9ef Statistics code cleanup
Summary: I'm separating code-cleanup part of https://reviews.facebook.net/D14517. This will make D14517 easier to understand and this diff easier to review.

Test Plan: make check

Reviewers: haobo, kailiu, sdong, dhruba, tnovak

Reviewed By: tnovak

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15099
2014-01-17 12:46:06 -08:00
Naman Gupta
1447bb5919 Allow callback to change size of existing value. Change return type of the callback function to an enum status to handle 3 cases.
Summary:
This diff fixes 2 hacks:
* The callback function can modify the existing value inplace, if the merged value fits within the existing buffer size. But currently the existing buffer size is not being modified. Now the callback recieves a int* allowing the size to be modified. Since size is encoded as a varint in the internal key for memtable. It might happen that the entire value might have be copied to the new location if the new size varint is smaller than the existing size varint.
* The callback function has 3 functionalities
    1. Modify existing buffer inplace, and update size correspondingly. Now to indicate that, Returns 1.
    2. Generate a new buffer indicating merged value. Returns 2.
    3. Fails to do either of above, based on whatever application logic. Returns 0.

Test Plan: Just make all for now. I'm adding another unit test to test each scenario.

Reviewers: dhruba, haobo

Reviewed By: haobo

CC: leveldb, sdong, kailiu, xinyaohu, sumeet, danguo

Differential Revision: https://reviews.facebook.net/D15195
2014-01-16 15:12:39 -08:00
kailiu
1304d8c8ce Merge branch 'master' into performance
Conflicts:
	Makefile
	db/db_impl.cc
	db/db_impl.h
	db/db_test.cc
	db/memtable.cc
	db/memtable.h
	db/version_edit.h
	db/version_set.cc
	include/rocksdb/options.h
	util/hash_skiplist_rep.cc
	util/options.cc
2014-01-15 23:12:31 -08:00
kailiu
eae1804f29 Remove the unnecessary use of shared_ptr
Summary:
shared_ptr is slower than unique_ptr (which literally comes with no performance cost compare with raw pointers).
In memtable and memtable rep, we use shared_ptr when we'd actually should use unique_ptr.

According to igor's previous work, we are likely to make quite some performance gain from this diff.

Test Plan: make check

Reviewers: dhruba, igor, sdong, haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15213
2014-01-15 18:22:01 -08:00
kailiu
c8f16221ed Fix the return type of WriteBatch::Data().
Summary: Quick fix for https://reviews.facebook.net/D15123

Test Plan: Make check

Reviewers: sdong, vkrest

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15165
2014-01-14 20:24:48 -08:00
Igor Canadi
d9cd7a063f Fix CompactRange to apply filter to every key
Summary:
When doing CompactRange(), we should first flush the memtable and then calculate max_level_with_files. Also, we want to compact all the levels that have files, including level `max_level_with_files`.

This patch fixed the unit test.

Test Plan: Added a failing unit test and a fix, so it's not failing anymore.

Reviewers: dhruba, haobo, sdong

Reviewed By: haobo

CC: leveldb, xjin

Differential Revision: https://reviews.facebook.net/D14421
2014-01-14 16:19:09 -08:00
Siying Dong
9ea8bf90f1 DB::Put() to estimate write batch data size needed and pre-allocate buffer
Summary:
In one of CPU profiles, we see some CPU costs of string::reserve() inside Batch.Put(). This patch should be able to reduce some of the costs by allocating sufficient buffer before hand.

Since it is a trivial percentage of CPU costs, I didn't find a way to show the improvement in one of the benchmarks. I'll deploy it to same application and do the same CPU profiling to make sure those CPU costs are reduced.

Test Plan: make all check

Reviewers: haobo, kailiu, igor

Reviewed By: haobo

CC: leveldb, nkg-

Differential Revision: https://reviews.facebook.net/D15135
2014-01-14 11:24:43 -08:00
Siying Dong
51dd21926c DB::Put() to estimate write batch data size needed and pre-allocate buffer
Summary:
In one of CPU profiles, we see some CPU costs of string::reserve() inside Batch.Put(). This patch should be able to reduce some of the costs by allocating sufficient buffer before hand.

Since it is a trivial percentage of CPU costs, I didn't find a way to show the improvement in one of the benchmarks. I'll deploy it to same application and do the same CPU profiling to make sure those CPU costs are reduced.

Test Plan: make all check

Reviewers: haobo, kailiu, igor

Reviewed By: haobo

CC: leveldb, nkg-

Differential Revision: https://reviews.facebook.net/D15135
2014-01-14 10:53:16 -08:00
Naman Gupta
8454cfe569 Add read/modify/write functionality to Put() api
Summary: The application can set a callback function, which is applied on the previous value. And calculates the new value. This new value can be set, either inplace, if the previous value existed in memtable, and new value is smaller than previous value. Otherwise the new value is added normally.

Test Plan: fbmake. Added unit tests. All unit tests pass.

Reviewers: dhruba, haobo

Reviewed By: haobo

CC: sdong, kailiu, xinyaohu, sumeet, leveldb

Differential Revision: https://reviews.facebook.net/D14745
2014-01-14 07:55:16 -08:00
Siying Dong
c4548d5f1f WriteBatch to provide a way for user to query data size directly and only return constant reference of data in Data()
Summary:
WriteBatch::Data() now is easily to be misuse by users. Also, there is no cheap way for user of WriteBatch to know the data size accumulated. This patch fix the problem by:
(1) return a constant reference to Data() so it's obvious to caller what it means.
(2) add a function to return data size directly

Test Plan: make all check

Reviewers: haobo, igor, kailiu

Reviewed By: kailiu

CC: zshao, leveldb

Differential Revision: https://reviews.facebook.net/D15123
2014-01-13 16:52:14 -08:00
Schalk-Willem Kruger
a09ee1069d Improve RocksDB "get" performance by computing merge result in memtable
Summary:
Added an option (max_successive_merges) that can be used to specify the
maximum number of successive merge operations on a key in the memtable.
This can be used to improve performance of the "get" operation. If many
successive merge operations are performed on a key, the performance of "get"
operations on the key deteriorates, as the value has to be computed for each
"get" operation by applying all the successive merge operations.

FB Task ID: #3428853

Test Plan:
make all check
db_bench --benchmarks=readrandommergerandom
counter_stress_test

Reviewers: haobo, vamsi, dhruba, sdong

Reviewed By: haobo

CC: zshao

Differential Revision: https://reviews.facebook.net/D14991
2014-01-10 17:33:56 -08:00
Siying Dong
aa0ef6602d [Performance Branch] If options.max_open_files set to be -1, cache table readers in FileMetadata for Get() and NewIterator()
Summary:
In some use cases, table readers for all live files should always be cached. In that case, there will be an opportunity to avoid the table cache look-up while Get() and NewIterator().

We define options.max_open_files = -1 to be the mode that table readers for live files will always be kept. In that mode, table readers are cached in FileMetaData (with a reference count hold in table cache). So that when executing table_cache.Get() and table_cache.newInterator(), LRU cache checking can be by-passed, to reduce latency.

Test Plan: add a test case in db_test

Reviewers: haobo, kailiu

Reviewed By: haobo

CC: dhruba, igor, leveldb

Differential Revision: https://reviews.facebook.net/D15039
2014-01-10 15:57:49 -08:00
Siying Dong
424a524ac9 [Performance Branch] A Hashed Linked List Based Mem Table
Summary:
Implement a mem table, in which keys are hashed based on prefixes. In each bucket, entries are organized in a sorted linked list. It has the same thread safety guarantee as skip list.

The motivation is to optimize memory usage for the case that prefix hashing is primary way of seeking to the entry. Compared to hash skip list implementation, this implementation is more memory efficient, but inside each bucket, search is always linear. The target scenario is that there are only very limited number of records in each hash bucket.

Test Plan: Add a test case in db_test

Reviewers: haobo, kailiu, dhruba

Reviewed By: haobo

CC: igor, nkg-, leveldb

Differential Revision: https://reviews.facebook.net/D14979
2014-01-09 16:19:11 -08:00
Igor Canadi
cb37ddf229 Feature requests for BackupableDB
Summary:
This diff introduces some features that were requested by two internal customers:
* Ability for backups not to share table files, because we can't guarantee that equal filename means equal content accross replicas
* Ability for two threads to call EnableFileDeletions() and DisableFileDeletions()
* Ability to stop backup from another thread and not slow down the DB close
* Copy the files to the temporary folder first and then atomically rename

Test Plan: Added some tests to backupable_db_test

Reviewers: dhruba, sanketh, muthu, sdong, haobo

Reviewed By: haobo

CC: leveldb, sanketh, muthu

Differential Revision: https://reviews.facebook.net/D14769
2014-01-09 12:24:28 -08:00
kailiu
12b6d2b839 Separate the aligned and unaligned memory allocation
Summary: Use two vectors for different types of memory allocation.

Test Plan: run all unit tests.

Reviewers: haobo, sdong

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D15027
2014-01-08 15:11:42 -08:00
Igor Canadi
a45b7d83ba Merge pull request #59 from mlin/more-c-bindings
C API: add rocksdb_env_set_high_priority_background_threads
2014-01-07 16:33:03 -08:00
kailiu
e72aa37cc5 Merge branch 'master' into performance
Conflicts:
	db/table_cache.cc
2014-01-02 16:34:59 -08:00
Igor Canadi
b60c14f6ee Support multi-threaded DisableFileDeletions() and EnableFileDeletions()
Summary:
We don't want two threads to clash if they concurrently call DisableFileDeletions() and EnableFileDeletions(). I'm adding a counter that will enable file deletions only after all DisableFileDeletions() calls have been negated with EnableFileDeletions().

However, we also don't want to break the old behavior, so I added a parameter force to EnableFileDeletions(). If force is true, we will still enable file deletions after every call to EnableFileDeletions(), which is what is happening now.

Test Plan: make check

Reviewers: dhruba, haobo, sanketh

Reviewed By: dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D14781
2014-01-02 03:33:42 -08:00
Mike Lin
4b1d049236 C API: add rocksdb_env_set_high_priority_background_threads 2013-12-31 15:14:18 -08:00
kailiu
f1cec73a76 Merge branch 'master' into performance
Conflicts:
	db/db_impl.cc
	db/db_test.cc
	db/memtable.cc
	db/version_set.cc
	include/rocksdb/statistics.h
2013-12-27 12:23:17 -08:00