Commit Graph

73 Commits

Author SHA1 Message Date
Aaron Gao
972f96b3fb direct io write support
Summary:
rocksdb direct io support

```
[gzh@dev11575.prn2 ~/rocksdb] ./db_bench -benchmarks=fillseq --num=1000000
Initializing RocksDB Options from the specified file
Initializing RocksDB Options from command-line flags
RocksDB:    version 5.0
Date:       Wed Nov 23 13:17:43 2016
CPU:        40 * Intel(R) Xeon(R) CPU E5-2660 v2 @ 2.20GHz
CPUCache:   25600 KB
Keys:       16 bytes each
Values:     100 bytes each (50 bytes after compression)
Entries:    1000000
Prefix:    0 bytes
Keys per prefix:    0
RawSize:    110.6 MB (estimated)
FileSize:   62.9 MB (estimated)
Write rate: 0 bytes/second
Compression: Snappy
Memtablerep: skip_list
Perf Level: 1
WARNING: Assertions are enabled; benchmarks unnecessarily slow
------------------------------------------------
Initializing RocksDB Options from the specified file
Initializing RocksDB Options from command-line flags
DB path: [/tmp/rocksdbtest-112628/dbbench]
fillseq      :       4.393 micros/op 227639 ops/sec;   25.2 MB/s

[gzh@dev11575.prn2 ~/roc
Closes https://github.com/facebook/rocksdb/pull/1564

Differential Revision: D4241093

Pulled By: lightmark

fbshipit-source-id: 98c29e3
2016-12-22 13:09:19 -08:00
Andrew Kryczka
f0c509e2c8 Return finer-granularity status from Env::GetChildren*
Summary:
It'd be nice to use the error status type to distinguish
between user error and system error. For example, GetChildren can fail
listing a backup directory's contents either because a bad path was provided
(user error) or because an operation failed, e.g., a remote storage service
call failed (system error). In the former case, we want to continue and treat
the backup directory as empty; in the latter case, we want to immediately
propagate the error to the caller.

This diff uses NotFound to indicate user error and IOError to indicate
system error. Previously IOError indicated both.
Closes https://github.com/facebook/rocksdb/pull/1644

Differential Revision: D4312157

Pulled By: ajkr

fbshipit-source-id: 51b4f24
2016-12-12 12:54:13 -08:00
Edouard A
99c052a34f Fix integer overflow in GetL0ThresholdSpeedupCompaction (#1378) 2016-10-23 18:43:29 -07:00
Dmitri Smirnov
b9311aa65c Implement WinRandomRW file and improve code reuse (#1388) 2016-10-13 16:36:34 -07:00
Edouard A
66a91e2607 Add NoSpace subcode to IOError (#1320)
Add a sub code to distinguish "out of space" errors from regular I/O errors
2016-09-07 12:37:45 -07:00
Islam AbdelRahman
e9b2af87f8 Expose ThreadPool under include/rocksdb/threadpool.h
Summary:
This diff split ThreadPool to
-ThreadPool (abstract interface exposed in include/rocksdb/threadpool.h)
-ThreadPoolImpl (actual implementation in util/threadpool_imp.h)

This allow us to expose ThreadPool to the user so we can use it as an option later

Test Plan: existing unit tests

Reviewers: andrewkr, yiwu, yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D62085
2016-08-26 10:41:35 -07:00
sdong
f62fbd2c85 Handle overflow case of rate limiter's paramters
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
2016-05-27 16:15:28 -07:00
Aaron Orenstein
2073cf3775 Eliminate use of 'using namespace std'. Also remove a number of ADL references to std functions.
Summary: Reduce use of argument-dependent name lookup in RocksDB.

Test Plan: 'make check' passed.

Reviewers: andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D58203
2016-05-20 07:42:18 -07:00
Dmitri Smirnov
26adaad438 Split WinEnv into separate classes. (#1128)
For ease of reuse and customization as a library
  without wrapping.
  WinEnvThreads is a class for replacement.
  WintEnvIO is a class for reuse and behavior override.
  Added private virtual functions for custom override
  of fallocate pread for io classes.
2016-05-19 16:40:54 -07:00
Dmitri Smirnov
bac3be7c46 Fix build issue. (#1123)
Implement GetUniqueIdFromFile to support new tests and the feature.
2016-05-16 17:01:00 -07:00
Dmitri Smirnov
aab91b8d8f Use generic threadpool for Windows environment (#1120)
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.
2016-05-12 18:34:04 -07:00
Dmitri Smirnov
4ea6e051ee Fix multiple issues with WinMmapFile fo sequential writing (#1108)
make preallocation inline with other writable files
  make sure that we map no more than pre-allocated size.
2016-04-29 16:43:13 -07:00
PraveenSinghRao
e8115cea45 Revert "Use async file handle for better parallelism (#1049)" (#1105)
This reverts commit b54c347424.

Revert async file handle change as it causes failures with appveyor
2016-04-28 22:50:26 -07:00
Li Peng
6d4832a998 Merge pull request #1101 from flyd1005/wip-fix-typo
fix typos and remove duplicated words
2016-04-28 02:30:44 -07:00
PraveenSinghRao
b54c347424 Use async file handle for better parallelism (#1049) 2016-04-22 13:27:33 -07:00
Dmitri Smirnov
ee221d2de0 Introduce XPRESS compresssion on Windows. (#1081)
Comparable with Snappy on comp ratio.
  Implemented using Windows API, does not require external package.
  Avaiable since Windows 8 and server 2012.
  Use -DXPRESS=1 with CMake to enable.
2016-04-19 22:54:24 -07:00
Yueh-Hsuan Chiang
a558830f8f Fixed compile warnings in posix_logger.h and coding.h
Summary:
Fixed the following compile warnings:

/Users/yhchiang/rocksdb/util/posix_logger.h:32:11: error: unused variable 'kDebugLogChunkSize' [-Werror,-Wunused-const-variable]
const int kDebugLogChunkSize = 128 * 1024;
          ^
/Users/yhchiang/rocksdb/util/coding.h:24:20: error: unused variable 'kMaxVarint32Length' [-Werror,-Wunused-const-variable]
const unsigned int kMaxVarint32Length = 5;
                   ^
2 errors generated.

Test Plan: make clean rocksdb

Reviewers: igor, sdong, anthony, IslamAbdelRahman, rven, kradhakrishnan, adamretter

Reviewed By: adamretter

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D56223
2016-03-31 16:01:47 -07:00
Dmitri Smirnov
2ca0994cf7 Latest versions of Jemalloc library do not require je_init()/je_unint()
calls. #ifdef in the source code and make this a default build option.
2016-03-17 11:25:20 -07:00
Dmitri Smirnov
9ea2968d26 Implement ConsistentChildrenAttribute
by using default implementation for now as it works.
2016-02-19 14:20:34 -08:00
Andrew Kryczka
d733dd5728 [build] Fix env_win.cc compiler errors
Summary: I broke it in D53781.

Test Plan: tried the same code in util/env_posix.cc and it compiled successfully

Reviewers: sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54303
2016-02-17 11:57:04 -08:00
Baraa Hamodi
21e95811d1 Updated all copyright headers to the new format. 2016-02-09 15:12:00 -08:00
Andrew Kryczka
59b3ee658f Env function for bulk metadata retrieval
Summary:
Added this new function, which returns filename, size, and modified
timestamp for each file in the provided directory. The default implementation
retrieves the metadata sequentially using existing functions. In the next diff
I'll make HdfsEnv override this function to use libhdfs's bulk get function.

This won't work on windows due to the path separator.

Test Plan:
new unit test

  $ ./env_test --gtest_filter=EnvPosixTest.ConsistentChildrenMetadata

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: IslamAbdelRahman, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53781
2016-02-09 14:54:32 -08:00
Tomas Kolda
57a95a7001 Making use of GetSystemTimePreciseAsFileTime dynamic - code review fixes 2016-02-02 10:23:56 +01:00
Tomas Kolda
502d41f150 Making use of GetSystemTimePreciseAsFileTime dynamic to not
break compatibility with Windows 7. The issue with rotated logs
was fixed other way.
2016-02-02 10:23:56 +01:00
Dmitri Smirnov
36300fbbe3 Enable per-request buffer allocation in RandomAccessFile
This change impacts only non-buffered I/O on Windows.
 Currently, there is a buffer per RandomAccessFile
 instance that is protected by a lock. The reason we
 maintain the buffer is non-buffered I/O requires an aligned
 buffer to work.
 XPerf traces demonstrate that we accumulate a considerable
 wait time while waiting for that lock.
 This change enables to set random access buffer size to zero
 which would indicate a per request allocation.
 We are expecting that allocation expense would be much less than
 I/O costs plus wait time due to the fact that the memory heap
 would tend to re-use page aligned allocations especially with the
 use of Jemalloc.
 This change does not affect buffer use as a read_ahead_buffer for
 compaction purposes.
2016-02-01 13:14:37 -08:00
Dmitri Smirnov
ac50fd3a71 Align statistics
Use Yield macro to make it a little more portable between platforms.
2016-01-13 14:53:23 -08:00
Marek Kurdej
92d0850f1c Fix failing assertion in logger on Windows when the disk is full. 2016-01-05 13:35:14 +01:00
Nathan Bronson
7d87f02799 support for concurrent adds to memtable
Summary:
This diff adds support for concurrent adds to the skiplist memtable
implementations.  Memory allocation is made thread-safe by the addition of
a spinlock, with small per-core buffers to avoid contention.  Concurrent
memtable writes are made via an additional method and don't impose a
performance overhead on the non-concurrent case, so parallelism can be
selected on a per-batch basis.

Write thread synchronization is an increasing bottleneck for higher levels
of concurrency, so this diff adds --enable_write_thread_adaptive_yield
(default off).  This feature causes threads joining a write batch
group to spin for a short time (default 100 usec) using sched_yield,
rather than going to sleep on a mutex.  If the timing of the yield calls
indicates that another thread has actually run during the yield then
spinning is avoided.  This option improves performance for concurrent
situations even without parallel adds, although it has the potential to
increase CPU usage (and the heuristic adaptation is not yet mature).

Parallel writes are not currently compatible with
inplace updates, update callbacks, or delete filtering.
Enable it with --allow_concurrent_memtable_write (and
--enable_write_thread_adaptive_yield).  Parallel memtable writes
are performance neutral when there is no actual parallelism, and in
my experiments (SSD server-class Linux and varying contention and key
sizes for fillrandom) they are always a performance win when there is
more than one thread.

Statistics are updated earlier in the write path, dropping the number
of DB mutex acquisitions from 2 to 1 for almost all cases.

This diff was motivated and inspired by Yahoo's cLSM work.  It is more
conservative than cLSM: RocksDB's write batch group leader role is
preserved (along with all of the existing flush and write throttling
logic) and concurrent writers are blocked until all memtable insertions
have completed and the sequence number has been advanced, to preserve
linearizability.

My test config is "db_bench -benchmarks=fillrandom -threads=$T
-batch_size=1 -memtablerep=skip_list -value_size=100 --num=1000000/$T
-level0_slowdown_writes_trigger=9999 -level0_stop_writes_trigger=9999
-disable_auto_compactions --max_write_buffer_number=8
-max_background_flushes=8 --disable_wal --write_buffer_size=160000000
--block_size=16384 --allow_concurrent_memtable_write" on a two-socket
Xeon E5-2660 @ 2.2Ghz with lots of memory and an SSD hard drive.  With 1
thread I get ~440Kops/sec.  Peak performance for 1 socket (numactl
-N1) is slightly more than 1Mops/sec, at 16 threads.  Peak performance
across both sockets happens at 30 threads, and is ~900Kops/sec, although
with fewer threads there is less performance loss when the system has
background work.

Test Plan:
1. concurrent stress tests for InlineSkipList and DynamicBloom
2. make clean; make check
3. make clean; DISABLE_JEMALLOC=1 make valgrind_check; valgrind db_bench
4. make clean; COMPILE_WITH_TSAN=1 make all check; db_bench
5. make clean; COMPILE_WITH_ASAN=1 make all check; db_bench
6. make clean; OPT=-DROCKSDB_LITE make check
7. verify no perf regressions when disabled

Reviewers: igor, sdong

Reviewed By: sdong

Subscribers: MarkCallaghan, IslamAbdelRahman, anthony, yhchiang, rven, sdong, guyg8, kradhakrishnan, dhruba

Differential Revision: https://reviews.facebook.net/D50589
2015-12-25 11:03:40 -08:00
Siying Dong
298ba27ae2 Merge pull request #846 from yuslepukhin/enble_c4244_lossofdata
Enable MS compiler warning c4244.
2015-12-23 22:59:42 -08:00
Venkatesh Radhakrishnan
030215bf01 Running manual compactions in parallel with other automatic or manual compactions in restricted cases
Summary:
This diff provides a framework for doing manual
compactions in parallel with other compactions. We now have a deque of manual compactions. We also pass manual compactions as an argument from RunManualCompactions down to
BackgroundCompactions, so that RunManualCompactions can be reentrant.
Parallelism is controlled by the two routines
ConflictingManualCompaction to allow/disallow new parallel/manual
compactions based on already existing ManualCompactions. In this diff, by default manual compactions still have to run exclusive of other compactions. However, by setting the compaction option, exclusive_manual_compaction to false, it is possible to run other compactions in parallel with a manual compaction. However, we are still restricted to one manual compaction per column family at a time. All of these restrictions will be relaxed in future diffs.
I will be adding more tests later.

Test Plan: Rocksdb regression + new tests + valgrind

Reviewers: igor, anthony, IslamAbdelRahman, kradhakrishnan, yhchiang, sdong

Reviewed By: sdong

Subscribers: yoshinorim, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D47973
2015-12-14 11:20:34 -08:00
Dmitri Smirnov
236fe21c92 Enable MS compiler warning c4244.
Mostly due to the fact that there are differences in sizes of int,long
  on 64 bit systems vs GNU.
2015-12-11 16:47:34 -08:00
Siying Dong
fa3dbf203f Merge pull request #853 from Vaisman/enable_C4267_warning
Enable C4267 warning
2015-12-08 17:59:24 -08:00
yuslepukhin
78de0c9222 Fix up VS 15 build.
Fix warnings
 Take advantage of native snprintf on VS 15
2015-12-08 08:38:21 -08:00
Vasili Svirski
41b32c6059 Enable C4267 warning
* conversion from 'size_t' to 'type', by add static_cast

Tested:
* by build solution on Windows, Linux locally,
* run tests
* build CI system successful
2015-11-24 16:33:09 +03:00
yuslepukhin
047bd22aae Build on Visual Studio 2015 Update 1 2015-11-20 15:31:47 -08:00
Dmitri Smirnov
314f62194a Remove headers from the cc since they are in the module's header. 2015-11-16 15:08:11 -08:00
Dmitri Smirnov
472c74006f Add necessary headers after cpplint rearranged includes 2015-11-16 14:41:11 -08:00
Islam AbdelRahman
a163cc2d5a Lint everything
Summary:
```
arc2 lint --everything
```

run the linter on the whole code repo to fix exisitng lint issues

Test Plan: make check -j64

Reviewers: sdong, rven, anthony, kradhakrishnan, yhchiang

Reviewed By: yhchiang

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D50769
2015-11-16 12:56:21 -08:00
Dmitri Smirnov
5270b33bd3 Make use of portable uint64_t type to make possible file access
in 64-bit.

  Currently, a signed off_t type is being used for the following
  interfaces for both offset and the length in bytes:
  * `Allocate`
  * `RangeSync`

  On Linux `off_t` is automatically either 32 or 64-bit depending on
  the platform. On Windows it is always a 32-bit signed long which
  limits file access and in particular space pre-allocation
  to effectively 2 Gb.

  Proposal is to replace off_t with uint64_t as a portable type
  always access files with 64-bit interfaces.

  May need to modify posix code but lack resources to test it.
2015-11-10 17:03:42 -08:00
sdong
296c3a1f94 "make format" in some recent commits
Summary: Run "make format" for some recent commits.

Test Plan: Build and run tests

Reviewers: IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D49707
2015-10-29 17:11:14 -07:00
Dmitri Smirnov
6fbc4f9f3e Implement smart buffer management.
introduce a new DBOption random_access_max_buffer_size to limit
  the size of the random access buffer used for unbuffered access.
  Implement read ahead buffering when enabled.
  To that effect propagate compaction_readahead_size and the new option
  to the env options to make it available for the implementation.
  Add Hint() override so SetupForCompaction() call would call Hint()
  readahead can now be setup from both Hint() and EnableReadAhead()
  Add new option random_access_max_buffer_size support
  db_bench, options_helper to make it string parsable
  and the unit test.
2015-10-27 14:44:16 -07:00
sdong
e1a5ff857b Allow users to disable some kill points in db_stress
Summary:
Give a name for every kill point, and allow users to disable some kill points based on prefixes. The kill points can be passed by db_stress through a command line paramter. This provides a way for users to boost the chance of triggering low frequency kill points
This allow follow up changes in crash test scripts to improve crash test coverage.

Test Plan:
Manually run db_stress with variable values of --kill_random_test and --kill_prefix_blacklist. Like this:
 --kill_random_test=2 --kill_prefix_blacklist=Posix,WritableFileWriter::Append,WritableFileWriter::WriteBuffered,WritableFileWriter::Sync

Reviewers: igor, kradhakrishnan, rven, IslamAbdelRahman, yhchiang

Reviewed By: yhchiang

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D48735
2015-10-15 14:33:13 -07:00
Siying Dong
d662b8dab5 Merge pull request #766 from PraveenSinghRao/lockfix
move debug variable under ifndef NDEBUG
2015-10-14 10:07:17 -07:00
Praveen Rao
91c041e578 move debug variable under ifndef 2015-10-13 14:28:11 -07:00
Igor Canadi
0be50ed12c Merge pull request #763 from PraveenSinghRao/lockfix
Fixing mutex to not use unique_lock
2015-10-12 16:03:04 -07:00
Praveen Rao
a1d37602a0 Fixing mutex to not use unique_lock 2015-10-12 15:41:20 -07:00
Dmitri Smirnov
e95b703b7f Mmap reads should not return error if reading past file
Summary:
  This mirrors  https://reviews.facebook.net/D45645
  Currently, mmap returns IOError when user tries to read
  data past the end of the file. This diff changes the behavior.
  Now, we return just the bytes that we can, and report the size
  we returned via a Slice result. This is consistent with non-mmap
  behavior and also pread() system call.
2015-10-06 16:19:58 -07:00
Dmitri Smirnov
489a3e95d4 Re-work to support size_t max constant for 32/64-bit. 2015-09-22 10:34:21 -07:00
Dmitri Smirnov
ddc8b44998 Address code review comments both GH and internal
Fix compilation issues on GCC/CLANG
 Address Windows Release test build issues due to Sync
2015-09-11 17:36:48 -07:00
Dmitri Smirnov
30e82d5c41 Refactor to support file_reader_writer on Windows.
Summary. A change https://reviews.facebook.net/differential/diff/224721/
  Has attempted to move common functionality out of platform dependent
  code to a new facility called file_reader_writer.
  This includes:
  - perf counters
  - Buffering
  - RateLimiting

  However, the change did not attempt to refactor Windows code.
  To mitigate, we introduce new quering interfaces such as UseOSBuffer(),
  GetRequiredBufferAlignment() and ReaderWriterForward()
  for pure forwarding where required.
  Introduce WritableFile got a new method Truncate(). This is to communicate
  to the file as to how much data it has on close.
   - When space is pre-allocated on Linux it is filled with zeros implicitly,
    no such thing exist on Windows so we must truncate file on close.
   - When operating in unbuffered mode the last page is filled with zeros but we still want to truncate.

   Previously, Close() would take care of it but now buffer management is shifted to the wrappers and the file has
   no idea about the file true size.

   This means that Close() on the wrapper level must always include
   Truncate() as well as wrapper __dtor should call Close() and
   against double Close().
   Move buffered/unbuffered write logic to the wrapper.
   Utilize Aligned buffer class.
   Adjust tests and implement Truncate() where necessary.
   Come up with reasonable defaults for new virtual interfaces.
   Forward calls for RandomAccessReadAhead class to avoid double
   buffering and locking (double locking in unbuffered mode on WIndows).
2015-09-11 09:57:02 -07:00