Commit Graph

13 Commits

Author SHA1 Message Date
Sagar Vemuri
440621aab8 Fix Copying of data between buffers in FilePrefetchBuffer (#4100)
Summary:
Copy data between buffers inside FilePrefetchBuffer only when chunk length is greater than 0. Otherwise AlignedBuffer was accessing memory out of its range causing crashes.

Removing the tracking of buffer length outside of `AlignedBuffer`, i.e. in `FilePrefetchBuffer` and `ReadaheadRandomAccessFile`, will follow in a separate PR, as it is not the root cause of the crash reported in #4051. (`FilePrefetchBuffer` itself has been this way from its inception, and `ReadaheadRandomAccessFile` was updated to add the buffer length at some point).

Comprehensive tests for `FilePrefetchBuffer` also to follow in a separate PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4100

Differential Revision: D8792590

Pulled By: sagar0

fbshipit-source-id: 3578f45761cf6884243e767f749db4016ccc93e1
2018-07-11 12:28:13 -07:00
Sagar Vemuri
7103559f49 Improve direct IO range scan performance with readahead (#3884)
Summary:
This PR extends the improvements in #3282 to also work when using Direct IO.
We see **4.5X performance improvement** in seekrandom benchmark doing long range scans, when using direct reads, on flash.

**Description:**
This change improves the performance of iterators doing long range scans (e.g. big/full index or table scans in MyRocks) by using readahead and prefetching additional data on each disk IO, and storing in a local buffer. This prefetching is automatically enabled on noticing more than 2 IOs for the same table file during iteration. The readahead size starts with 8KB and is exponentially increased on each additional sequential IO, up to a max of 256 KB. This helps in cutting down the number of IOs needed to complete the range scan.

**Implementation Details:**
- Used `FilePrefetchBuffer` as the underlying buffer to store the readahead data. `FilePrefetchBuffer` can now take file_reader, readahead_size and max_readahead_size as input to the constructor, and automatically do readahead.
- `FilePrefetchBuffer::TryReadFromCache` can now call `FilePrefetchBuffer::Prefetch` if readahead is enabled.
- `AlignedBuffer` (which is the underlying store for `FilePrefetchBuffer`) now takes a few additional args in `AlignedBuffer::AllocateNewBuffer` to allow copying data from the old buffer.
- Made sure not to re-read partial chunks of data that were already available in the buffer, from device again.
- Fixed a couple of cases where `AlignedBuffer::cursize_` was not being properly kept up-to-date.

**Constraints:**
- Similar to #3282, this gets currently enabled only when ReadOptions.readahead_size = 0 (which is the default value).
- Since the prefetched data is stored in a temporary buffer allocated on heap, this could increase the memory usage if you have many iterators doing long range scans simultaneously.
- Enabled only for user reads, and disabled for compactions. Compaction reads are controlled by the options `use_direct_io_for_flush_and_compaction` and `compaction_readahead_size`, and the current feature takes precautions not to mess with them.

**Benchmarks:**
I used the same benchmark as used in #3282.
Data fill:
```
TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=fillrandom -num=1000000000 -compression_type="none" -level_compaction_dynamic_level_bytes
```

Do a long range scan: Seekrandom with large number of nexts
```
TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=seekrandom -use_direct_reads -duration=60 -num=1000000000 -use_existing_db -seek_nexts=10000 -statistics -histogram
```

```
Before:
seekrandom   :   37939.906 micros/op 26 ops/sec;   29.2 MB/s (1636 of 1999 found)
With this change:
seekrandom   :   8527.720 micros/op 117 ops/sec;  129.7 MB/s (6530 of 7999 found)
```
~4.5X perf improvement. Taken on an average of 3 runs.
Closes https://github.com/facebook/rocksdb/pull/3884

Differential Revision: D8082143

Pulled By: sagar0

fbshipit-source-id: 4d7a8561cbac03478663713df4d31ad2620253bb
2018-06-21 11:13:08 -07:00
Anand Ananthabhotla
f9f4d40f93 Align SST file data blocks to avoid spanning multiple pages
Summary:
Provide a block_align option in BlockBasedTableOptions to allow
alignment of SST file data blocks. This will avoid higher
IOPS/throughput load due to < 4KB data blocks spanning 2 4KB pages.
When this option is set to true, the block alignment is set to lower of
block size and 4KB.
Closes https://github.com/facebook/rocksdb/pull/3502

Differential Revision: D7400897

Pulled By: anand1976

fbshipit-source-id: 04cc3bd144e88e3431a4f97604e63ad7a0f06d44
2018-03-26 20:26:10 -07:00
Maysam Yabandeh
4927b4e662 Rounddown in FilePrefetchBuffer::Prefetch
Summary:
FilePrefetchBuffer::Prefetch is currently rounds the offset up which does not fit its new use cases in prefetching index/filter blocks, as it would skips over some the offsets that were requested to be prefetched. This patch rounds down instead.

Fixes #3180
Closes https://github.com/facebook/rocksdb/pull/3413

Differential Revision: D6816392

Pulled By: maysamyabandeh

fbshipit-source-id: 3aaeaf59c55d72b61dacfae6d4a8e65eccb3c553
2018-01-26 12:57:25 -08:00
Siying Dong
3c327ac2d0 Change RocksDB License
Summary: Closes https://github.com/facebook/rocksdb/pull/2589

Differential Revision: D5431502

Pulled By: siying

fbshipit-source-id: 8ebf8c87883daa9daa54b2303d11ce01ab1f6f75
2017-07-15 16:11:23 -07:00
Siying Dong
0175d58c3c Make direct I/O write use incremental buffer
Summary:
Currently for direct I/O, the large maximum buffer is always allocated. This will be wasteful if users flush the data in much smaller chunks. This diff fix this by changing the behavior of incremental buffer works. When we enlarge buffer, we try to copy the existing data in the buffer to the enlarged buffer, rather than flush the buffer first. This can make sure that no extra I/O is introduced because of buffer enlargement.
Closes https://github.com/facebook/rocksdb/pull/2403

Differential Revision: D5178403

Pulled By: siying

fbshipit-source-id: a8fe1e7304bdb8cab2973340022fe80ff83449fd
2017-06-13 04:41:37 -07:00
Siying Dong
d616ebea23 Add GPLv2 as an alternative license.
Summary: Closes https://github.com/facebook/rocksdb/pull/2226

Differential Revision: D4967547

Pulled By: siying

fbshipit-source-id: dd3b58ae1e7a106ab6bb6f37ab5c88575b125ab4
2017-04-27 18:06:12 -07:00
Dmitri Smirnov
cdad04b051 Remove double buffering on RandomRead on Windows.
Summary:
Remove double buffering on RandomRead on Windows.
  With more logic appear in file reader/write Read no longer
  obeys forwarding calls to Windows implementation.
  Previously direct_io (unbuffered) was only available on Windows
  but now is supported as generic.
  We remove intermediate buffering on Windows.
  Remove random_access_max_buffer_size option which was windows specific.
  Non-zero values for that opton introduced unnecessary lock contention.
  Remove Env::EnableReadAhead(), Env::ShouldForwardRawRequest() that are
  no longer necessary.
  Add aligned buffer reads for cases when requested reads exceed read ahead size.
Closes https://github.com/facebook/rocksdb/pull/2105

Differential Revision: D4847770

Pulled By: siying

fbshipit-source-id: 8ab48f8e854ab498a4fd398a6934859792a2788f
2017-04-27 12:30:05 -07:00
Aaron Gao
dc2584eea0 direct reads refactor
Summary:
direct IO reads refactoring
remove unnecessary classes and unified interfaces
tested with db_bench

need more change for options and ON/OFF for different files.
Since disabled is default, it should be fine now
Closes https://github.com/facebook/rocksdb/pull/1636

Differential Revision: D4307189

Pulled By: lightmark

fbshipit-source-id: 6991e22
2017-01-11 16:54:12 -08:00
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
Baraa Hamodi
21e95811d1 Updated all copyright headers to the new format. 2016-02-09 15:12:00 -08:00
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