Skip to content

R
rocksdb

kvdb / rocksdb
12 个月 前同步成功

通知 0
Star 0
Fork 0
R
rocksdb

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 17 6月, 2016 2 次提交
  3. 16 6月, 2016 3 次提交
  5. 15 6月, 2016 6 次提交
    • I
      Fix clang build · 8366e10f
      由 Islam AbdelRahman 提交于 
      Summary: Fix clang build

Test Plan: USE_CLANG=1 make check -j64

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59667
      8366e10f
    • S
      Move away from enum char value -1 · 0babce57
      由 sdong 提交于 
      Summary: char is not signed in some platforms. Having negative values confuse those compilers.

Test Plan: Run all existing tests.

Reviewers: andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59619
      0babce57
    • I
      Optimize BlockIter::Prev() by caching decoded entries · 812dbfb4
      由 Islam AbdelRahman 提交于 
      Summary:
Right now the way we do BlockIter::Prev() is like this

- Go to the beginning of the restart interval
- Keep moving forward (and decoding keys using ParseNextKey()) until we reach the desired key

This can be optimized by caching the decoded entries in the first pass and reusing them in consecutive BlockIter::Prev() calls

Before caching

```
DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readreverse" --db="/dev/shm/bench_prev_opt/" --use_existing_db --disable_auto_compactions
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.413 micros/op 2423972 ops/sec;  268.2 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.414 micros/op 2413867 ops/sec;  267.0 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.410 micros/op 2440881 ops/sec;  270.0 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.414 micros/op 2417298 ops/sec;  267.4 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.413 micros/op 2421682 ops/sec;  267.9 MB/s
```

After caching

```
DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readreverse" --db="/dev/shm/bench_prev_opt/" --use_existing_db --disable_auto_compactions
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.324 micros/op 3088955 ops/sec;  341.7 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.335 micros/op 2980999 ops/sec;  329.8 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.341 micros/op 2929681 ops/sec;  324.1 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.344 micros/op 2908490 ops/sec;  321.8 MB/s
DB path: [/dev/shm/bench_prev_opt/]
readreverse  :       0.338 micros/op 2958404 ops/sec;  327.3 MB/s
```

Test Plan: COMPILE_WITH_ASAN=1 make check -j64

Reviewers: andrewkr, yiwu, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, yoshinorim

Differential Revision: https://reviews.facebook.net/D59463
      812dbfb4
    • K
      Minor bug fix with log name · 550bf895
      由 krad 提交于
      550bf895
    • I
      Fix examples/Makefile jemalloc error · 886af591
      由 Islam AbdelRahman 提交于 
      Summary: compile and link with jemalloc in examples binaries

Test Plan:
cd examples
make -j64

Reviewers: andrewkr, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, yoshinorim

Differential Revision: https://reviews.facebook.net/D59457
      886af591
    • A
      Expose save points in Java WriteBatch and WBWI (#1092) · e3b1e3df
      由 Adam Retter 提交于 
      * Java API - Expose SetSavePoint and RollbackToRestorePoint for WriteBatch and WriteBatchWithIndex

* Minor cleanup
      e3b1e3df
  7. 14 6月, 2016 6 次提交
    • I
      Remove wasteful instrumentation in FullMerge (stacked on D59577) · f5177c76
      由 Islam AbdelRahman 提交于 
      Summary:
[ This diff is stacked on top of D59577 ]

We keep calling timer.ElapsedNanos() on every call to MergeOperator::FullMerge even when statistics are disabled, this is wasteful.

I run the readseq benchmark on a DB containing 100K merge operands for 100K keys (1 operand per key) with 1GB block cache
I see slight performance improvment

Original results

```
$ ./db_bench --benchmarks="readseq,readseq,readseq,readseq,readseq,readseq,readseq,readseq,readseq" --merge_operator="max" --merge_keys=100000 --num=100000 --db="/dev/shm/100K_merge_compacted/" --cache_size=1073741824 --use_existing_db --disable_auto_compactions
------------------------------------------------
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.498 micros/op 2006597 ops/sec;  222.0 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.295 micros/op 3393627 ops/sec;  375.4 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.285 micros/op 3511155 ops/sec;  388.4 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.286 micros/op 3500470 ops/sec;  387.2 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.283 micros/op 3530751 ops/sec;  390.6 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.289 micros/op 3464811 ops/sec;  383.3 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.277 micros/op 3612814 ops/sec;  399.7 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.283 micros/op 3539640 ops/sec;  391.6 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.285 micros/op 3503766 ops/sec;  387.6 MB/s
```

After patch

```
$ ./db_bench --benchmarks="readseq,readseq,readseq,readseq,readseq,readseq,readseq,readseq,readseq" --merge_operator="max" --merge_keys=100000 --num=100000 --db="/dev/shm/100K_merge_compacted/" --cache_size=1073741824 --use_existing_db --disable_auto_compactions
------------------------------------------------
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.476 micros/op 2100119 ops/sec;  232.3 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.278 micros/op 3600887 ops/sec;  398.4 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.275 micros/op 3636698 ops/sec;  402.3 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.271 micros/op 3691661 ops/sec;  408.4 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.273 micros/op 3661534 ops/sec;  405.1 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.276 micros/op 3627106 ops/sec;  401.3 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.272 micros/op 3682635 ops/sec;  407.4 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.266 micros/op 3758331 ops/sec;  415.8 MB/s
DB path: [/dev/shm/100K_merge_compacted/]
readseq      :       0.266 micros/op 3761907 ops/sec;  416.2 MB/s
```

Test Plan: make check -j64

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59583
      f5177c76
    • G
      Remove dead Jenkins code and support `arc diff --preview` in RocksDB · 97fd2a63
      由 Gunnar Kudrjavets 提交于 
      Summary:
Two changes here:

- Remove dead Jenkins related code which is no longer relevant.
- Support `arc diff --preview`. Currently it doesn't work because a step which applies a diff assumes that a revision has been created. Which in case of `--preview` isn't. Therefore diff can't be applied and validation fails. Solution is to use `--nocommit` because for validation purposes performing a commit isn't necessary.

Test Plan:
- Current changes are submitted using `arc diff --preview`.
- All the pre-commit verification tests passed.

Reviewers: kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: leveldb, andrewkr, jtolmer, dhruba

Differential Revision: https://reviews.facebook.net/D59571
      97fd2a63
    • I
      Reuse TimedFullMerge instead of FullMerge + instrumentation · 7c919dec
      由 Islam AbdelRahman 提交于 
      Summary:
We have alot of code duplication whenever we call FullMerge we keep duplicating the instrumentation and statistics code
This is a simple diff to refactor the code to use TimedFullMerge instead of FullMerge

Test Plan: COMPILE_WITH_ASAN=1 make check -j64

Reviewers: andrewkr, yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59577
      7c919dec
    • S
      Remove the comments saying allow_concurrent_memtable_write and... · 9a33a723
      由 sdong 提交于 
      Remove the comments saying allow_concurrent_memtable_write and enable_write_thread_adaptive_yield are not stable

Summary: allow_concurrent_memtable_write and enable_write_thread_adaptive_yield went through extensive stress test and benchmarks. Should be considered stable.

Test Plan: N/A

Reviewers: ngbronson, yhchiang, IslamAbdelRahman, dhruba

Reviewed By: dhruba

Subscribers: MarkCallaghan, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59415
      9a33a723
    • K
      Fix tsan error · 81f6b33d
      由 krad 提交于 
      Summary:
The tsan error was because the random implementation we have is not
thread safe, using Random::GetTLSInstance

Test Plan: Run tests in Linux

Reviewers: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D59559
      81f6b33d
    • Y
      add option to not flush memtable on open() · bc8af90e
      由 Yi Wu 提交于 
      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
      bc8af90e
  9. 12 6月, 2016 1 次提交
  11. 11 6月, 2016 10 次提交
  13. 10 6月, 2016 8 次提交
  15. 09 6月, 2016 4 次提交
    • D
      netflix use-case · 00a05872
      由 dhruba borthakur 提交于
      00a05872
    • A
      use branch names in format compatibility test · 5091dfc1
      由 Andrew Kryczka 提交于 
      Summary:
We had to go back and update the g++ path for 4.4.fb-4.8.fb. So the
path is now fixed on the branches, but can't be fixed on the tags since they're
immutable. By making format compatibility tests use branch names (when
available), backported fixes like this will be used without having to re-release.

Also removed v1.5.7 and v2.1 because make fails.

Test Plan:
  $ build_tools/rocksdb-lego-determinator run_format_compatible

Reviewers: sdong, lightmark, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D59355
      5091dfc1
    • I
      Use valgrind built with gcc-4.9-glibc-2.20 · edc764e9
      由 Islam AbdelRahman 提交于 
      Summary:
Right now we use valgrind built with gcc-5-glibc-2.23 which keep generating this error everywhere

```
==3224756== Conditional jump or move depends on uninitialised value(s)
==3224756==    at 0x401AD68: index (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x4008933: _dl_map_object (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x4000ECC: map_doit (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x40101C7: _dl_catch_error (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x4000B34: do_preload (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x400458C: dl_main (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x40185DB: _dl_sysdep_start (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x4004F39: _dl_start (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==    by 0x4000CE7: ??? (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==  Uninitialised value was created by a stack allocation
==3224756==    at 0x4004519: dl_main (in /usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld-2.20.so)
==3224756==
```

Test Plan: DISABLE_JEMALLOC=1 make valgrind_check -j64

Reviewers: andrewkr, yiwu, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59343
      edc764e9
    • I
      Disable PersistentCacheTierTest.VolatileCacheInsertWithEviction test under TSAN temporarily · 8ff59b2b
      由 Islam AbdelRahman 提交于 
      Summary: Disable the test under TSAN temporary to temporarily the build

Test Plan: run the test under TSAN

Reviewers: kradhakrishnan, andrewkr, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59337
      8ff59b2b