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. 24 3月, 2017 1 次提交
    • W
      Add C API functions (and tests) for WriteBatchWithIndex · 41ccae6d
      由 Warren Falk 提交于 
      Summary:
I've added functions to the C API to support WriteBatchWithIndex as requested in #1833.

I've also added unit tests to c_test

I've implemented the WriteBatchWithIndex variation of every function available for regular WriteBatch.  And added additional functions unique to WriteBatchWithIndex.

For now, the following is omitted:
  1. The ability to create WriteBatchWithIndex's custom batch-only iterator as I'm not sure what its purpose is.  It should be possible to add later if anyone wants it.
  2. The ability to create the batch with a fallback comparator, since it appears to be unnecessary.  I believe the column family comparator will be used for this, meaning those using a custom comparator can just use the column family variations.
Closes https://github.com/facebook/rocksdb/pull/1985

Differential Revision: D4760039

Pulled By: siying

fbshipit-source-id: 393227e
      41ccae6d
  3. 23 3月, 2017 5 次提交
  5. 22 3月, 2017 1 次提交
  7. 21 3月, 2017 1 次提交
  9. 17 3月, 2017 2 次提交
    • I
      Break stalls when no bg work is happening · d52f334c
      由 Islam AbdelRahman 提交于 
      Summary:
Current stall will keep sleeping even if there is no Flush/Compactions to wait for, I changed the logic to break the stall if we are not flushing or compacting

db_bench command used
```
# fillrandom
# memtable size = 10MB
# value size = 1 MB
# num = 1000
# use /dev/shm
./db_bench --benchmarks="fillrandom,stats" --value_size=1048576 --write_buffer_size=10485760 --num=1000 --delayed_write_rate=XXXXX  --db="/dev/shm/new_stall" | grep "Cumulative stall"
```

```
Current results

# delayed_write_rate = 1000 Kb/sec
Cumulative stall: 00:00:9.031 H:M:S

# delayed_write_rate = 200 Kb/sec
Cumulative stall: 00:00:22.314 H:M:S

# delayed_write_rate = 100 Kb/sec
Cumulative stall: 00:00:42.784 H:M:S

# delayed_write_rate = 50 Kb/sec
Cumulative stall: 00:01:23.785 H:M:S

# delayed_write_rate = 25 Kb/sec
Cumulative stall: 00:02:45.702 H:M:S
```

```
New results

# delayed_write_rate = 1000 Kb/sec
Cumulative stall: 00:00:9.017 H:M:S

# delayed_write_rate = 200 Kb/sec
Cumulative stall: 00
Closes https://github.com/facebook/rocksdb/pull/1884

Differential Revision: D4585439

Pulled By: IslamAbdelRahman

fbshipit-source-id: aed2198
      d52f334c
    • I
      Support SstFileManager::SetDeleteRateBytesPerSecond() · 995618a8
      由 Islam AbdelRahman 提交于 
      Summary:
Update DeleteScheduler component to support changing delete rate in runtime by introducing
SstFileManager::SetDeleteRateBytesPerSecond()
Closes https://github.com/facebook/rocksdb/pull/1994

Differential Revision: D4719906

Pulled By: IslamAbdelRahman

fbshipit-source-id: e6b8d9e
      995618a8
  11. 16 3月, 2017 1 次提交
    • I
      Add macros to include file name and line number during Logging · e1916368
      由 Islam AbdelRahman 提交于 
      Summary:
current logging
```
2017/03/14-14:20:30.393432 7fedde9f5700 (Original Log Time 2017/03/14-14:20:30.393414) [default] Level summary: base level 1 max bytes base 268435456 files[1 0 0 0 0 0 0] max score 0.25
2017/03/14-14:20:30.393438 7fedde9f5700 [JOB 2] Try to delete WAL files size 61417909, prev total WAL file size 73820858, number of live WAL files 2.
2017/03/14-14:20:30.393464 7fedde9f5700 [DEBUG] [JOB 2] Delete /dev/shm/old_logging//MANIFEST-000001 type=3 #1 -- OK
2017/03/14-14:20:30.393472 7fedde9f5700 [DEBUG] [JOB 2] Delete /dev/shm/old_logging//000003.log type=0 #3 -- OK
2017/03/14-14:20:31.427103 7fedd49f1700 [default] New memtable created with log file: #9. Immutable memtables: 0.
2017/03/14-14:20:31.427179 7fedde9f5700 [JOB 3] Syncing log #6
2017/03/14-14:20:31.427190 7fedde9f5700 (Original Log Time 2017/03/14-14:20:31.427170) Calling FlushMemTableToOutputFile with column family [default], flush slots available 1, compaction slots allowed 1, compaction slots scheduled 1
2017/03/14-14:20:31.
Closes https://github.com/facebook/rocksdb/pull/1990

Differential Revision: D4708695

Pulled By: IslamAbdelRahman

fbshipit-source-id: cb8968f
      e1916368
  13. 14 3月, 2017 4 次提交
    • M
      Pinnableslice (2nd attempt) · 11526252
      由 Maysam Yabandeh 提交于 
      Summary:
PinnableSlice

    Summary:
    Currently the point lookup values are copied to a string provided by the
    user. This incures an extra memcpy cost. This patch allows doing point lookup
    via a PinnableSlice which pins the source memory location (instead of
    copying their content) and releases them after the content is consumed
    by the user. The old API of Get(string) is translated to the new API
    underneath.

    Here is the summary for improvements:

    value 100 byte: 1.8% regular, 1.2% merge values
    value 1k byte: 11.5% regular, 7.5% merge values
    value 10k byte: 26% regular, 29.9% merge values
    The improvement for merge could be more if we extend this approach to
    pin the merge output and delay the full merge operation until the user
    actually needs it. We have put that for future work.

    PS:
    Sometimes we observe a small decrease in performance when switching from
    t5452014 to this patch but with the old Get(string) API. The d
Closes https://github.com/facebook/rocksdb/pull/1756

Differential Revision: D4391738

Pulled By: maysamyabandeh

fbshipit-source-id: 6f3edd3
      11526252
    • S
      Add a new SstFileWriter constructor without explicit comparator · 1ffbdfd9
      由 Sagar Vemuri 提交于 
      Summary:
The comparator param in SstFileWriter constructor is redundant as it already exists as a field in options. So the current SstFileWriter constructor should be deprecated in favor of a new one which does not take a comparator.
Note that the jni/java apis have not been touched yet.
Closes https://github.com/facebook/rocksdb/pull/1978

Differential Revision: D4685629

Pulled By: sagar0

fbshipit-source-id: 372ce96
      1ffbdfd9
    • R
      Add ability to search for key prefix in sst_dump tool · ebd5639b
      由 Reid Horuff 提交于 
      Summary:
Add the flag --prefix to the sst_dump tool
This flag is similar to, and exclusive from, the --from flag.

--prefix=0x00FF will return all rows prefixed with 0x00FF.
The --to flag may also be specified and will work as expected.

These changes were used to help in debugging the power cycle corruption issue and theses changes were tested by scanning through a udb.
Closes https://github.com/facebook/rocksdb/pull/1984

Differential Revision: D4691814

Pulled By: reidHoruff

fbshipit-source-id: 027f261
      ebd5639b
    • M
      Fix some bugs in MockEnv · e6725e8c
      由 Maysam Yabandeh 提交于 
      Summary:
Fixing some bugs in MockEnv so it be actually used.
Closes https://github.com/facebook/rocksdb/pull/1914

Differential Revision: D4609923

Pulled By: maysamyabandeh

fbshipit-source-id: ca25735
      e6725e8c
  15. 09 3月, 2017 4 次提交
  17. 08 3月, 2017 2 次提交
    • S
      Add a memtable-only iterator · 97edc72d
      由 Sagar Vemuri 提交于 
      Summary:
This PR is to support a way to iterate over all the keys that are just in memtables.
Closes https://github.com/facebook/rocksdb/pull/1953

Differential Revision: D4663500

Pulled By: sagar0

fbshipit-source-id: 144e177
      97edc72d
    • L
      fix db_sst_test flakiness · 72202962
      由 Leonidas Galanis 提交于 
      Summary:
db_sst_test had been flaky occasionally in the following way: reached_max_space_on_compaction can in very rare cases be 0. This happens when the limit on maximum allowable space set using SetMaxAllowedSpaceUsage is hit during flush for all test db sizes (1,2,4,8 and 10MB).The fix clears the error returned when the the space limit is reached during flush. This ensures that the compaction call back will always be called. The runtime is increased slightly because the 1MB loop writes more data and hits the limit during multiple flushes until compaction is scheduled.
Closes https://github.com/facebook/rocksdb/pull/1861

Differential Revision: D4557396

Pulled By: lgalanis

fbshipit-source-id: ff778d1
      72202962
  19. 07 3月, 2017 1 次提交
    • R
      Set logs as getting flushed before releasing lock, race condition fix · 58b12dfe
      由 Reid Horuff 提交于 
      Summary:
Relating to #1903:

In MaybeFlushColumnFamilies() we want to modify the 'getting_flushed' flag before releasing the db mutex when SwitchMemtable() is called.

The following 2 actions need to be atomic in MaybeFlushColumnFamilies()
- getting_flushed is false on oldest log
- we determine that all CFs can be flushed to successfully release oldest log
- we set getting_flushed = true on the oldest log.
-------
- getting_flushed is false on oldest log
- we determine that all CFs can NOT be flushed to successfully release oldest log
- we set unable_to_flush_oldest_log_ = true on the oldest log.

#### In the 2pc case:

T1 enters function but is unable to flush all CFs to release log
T1 sets unable_to_flush_oldest_log_ = true
T1 begins flushing all CFs possible

T2 enters function but is unable to flush all CFs to release log
T2 sees unable_to_flush_oldes_log_ has been set so exits

T3 enters function and will be able to flush all CFs to release oldest log
T3 sets getting_flushed = true on oldes
Closes https://github.com/facebook/rocksdb/pull/1909

Differential Revision: D4646235

Pulled By: reidHoruff

fbshipit-source-id: c8d0447
      58b12dfe
  21. 06 3月, 2017 1 次提交
    • M
      Fix a bug in tests in options operator= · 534581a3
      由 Maysam Yabandeh 提交于 
      Summary:
Note: Using the default operator= is an unsafe approach for Options since it destructs shared_ptr in
the same order of their creation, in contrast to destructors which
destructs them in the opposite order of creation. One particular problme is
that the cache destructor might invoke callback functions that use Option
members such as statistics. To work around this problem, we manually call
destructor of table_facotry which eventually clears the block cache.
Closes https://github.com/facebook/rocksdb/pull/1950

Differential Revision: D4655473

Pulled By: maysamyabandeh

fbshipit-source-id: 6c4bbff
      534581a3
  23. 04 3月, 2017 1 次提交
  25. 03 3月, 2017 3 次提交
  27. 01 3月, 2017 3 次提交
  29. 28 2月, 2017 2 次提交
  31. 24 2月, 2017 4 次提交
  33. 23 2月, 2017 2 次提交
  35. 22 2月, 2017 2 次提交
    • M
      Fix interference between max_total_wal_size and db_write_buffer_size checks · 18eeb7b9
      由 Mike Kolupaev 提交于 
      Summary:
This is a trivial fix for OOMs we've seen a few days ago in logdevice.

RocksDB get into the following state:
(1) Write throughput is too high for flushes to keep up. Compactions are out of the picture - automatic compactions are disabled, and for manual compactions we don't care that much if they fall behind. We write to many CFs, with only a few L0 sst files in each, so compactions are not needed most of the time.
(2) total_log_size_ is consistently greater than GetMaxTotalWalSize(). It doesn't get smaller since flushes are falling ever further behind.
(3) Total size of memtables is way above db_write_buffer_size and keeps growing. But the write_buffer_manager_->ShouldFlush() is not checked because (2) prevents it (for no good reason, afaict; this is what this commit fixes).
(4) Every call to WriteImpl() hits the MaybeFlushColumnFamilies() path. This keeps flushing the memtables one by one in order of increasing log file number.
(5) No write stalling trigger is hit. We rely on max_write_buffer_number
Closes https://github.com/facebook/rocksdb/pull/1893

Differential Revision: D4593590

Pulled By: yiwu-arbug

fbshipit-source-id: af79c5f
      18eeb7b9
    • A
      level compaction expansion · 2a0f3d0d
      由 Aaron Gao 提交于 
      Summary:
reimplement the compaction expansion on lower level.

Considering such a case:
input level file: 1[B E] 2[F G] 3[H I] 4 [J M]
output level file: 5[A C] 6[D K] 7[L O]

If we initially pick file 2, now we will compact file 2 and 6. But we can safely compact 2, 3 and 6 without expanding the output level.

The previous code is messy and wrong.

In this diff, I first determine the input range [a, b], and output range [c, d],
then we get the range [e,f] = [min(a, c), max(b, d] and put all eligible clean-cut files within [e, f] into this compaction.

**Note: clean-cut means the files don't have the same user key on the boundaries of some files that are not chosen in this compaction**.
Closes https://github.com/facebook/rocksdb/pull/1760

Differential Revision: D4395564

Pulled By: lightmark

fbshipit-source-id: 2dc2c5c
      2a0f3d0d