Commit Graph

673 Commits

Author SHA1 Message Date
hyunwoo
0ebdd70579 fixed typo
Summary:
fixed typo
Closes https://github.com/facebook/rocksdb/pull/2312

Differential Revision: D5079631

Pulled By: sagar0

fbshipit-source-id: e4c8d1d89b244ee69e9dea1dd013227cc5241026
2017-05-17 16:41:49 -07:00
Aaron Gao
259a00eaca unbiase readamp bitmap
Summary:
Consider BlockReadAmpBitmap with bytes_per_bit = 32. Suppose bytes [a, b) were used, while bytes [a-32, a)
 and [b+1, b+33) weren't used; more formally, the union of ranges passed to BlockReadAmpBitmap::Mark() contains [a, b) and doesn't intersect with [a-32, a) and [b+1, b+33). Then bits [floor(a/32), ceil(b/32)] will be set, and so the number of useful bytes will be estimated as (ceil(b/32) - floor(a/32)) * 32, which is on average equal to b-a+31.

An extreme example: if we use 1 byte from each block, it'll be counted as 32 bytes from each block.

It's easy to remove this bias by slightly changing the semantics of the bitmap. Currently each bit represents a byte range [i*32, (i+1)*32).

This diff makes each bit represent a single byte: i*32 + X, where X is a random number in [0, 31] generated when bitmap is created. So, e.g., if you read a single byte at random, with probability 31/32 it won't be counted at all, and with probability 1/32 it will be counted as 32 bytes; so, on average it's counted as 1 byte.

*But there is one exception: the last bit will always set with the old way.*

(*) - assuming read_amp_bytes_per_bit = 32.
Closes https://github.com/facebook/rocksdb/pull/2259

Differential Revision: D5035652

Pulled By: lightmark

fbshipit-source-id: bd98b1b9b49fbe61f9e3781d07f624e3cbd92356
2017-05-10 01:49:52 -07:00
Islam AbdelRahman
4897eb250b dont skip IO for filter blocks
Summary:
Based on my experience with linkbench, We should not skip loading bloom filter blocks when they are not available in block cache when using Iterator::Seek

Actually I am not sure why this behavior existed in the first place
Closes https://github.com/facebook/rocksdb/pull/2255

Differential Revision: D5010721

Pulled By: maysamyabandeh

fbshipit-source-id: 0af545a06ac4baeecb248706ec34d009c2480ca4
2017-05-09 09:52:02 -07:00
Maysam Yabandeh
40af2381ec Object lifetime in cache
Summary:
Any non-raw-data dependent object must be destructed before the table
    closes. There was a bug of not doing that for filter object. This patch
    fixes the bug and adds a unit test to prevent such bugs in future.
Closes https://github.com/facebook/rocksdb/pull/2246

Differential Revision: D5001318

Pulled By: maysamyabandeh

fbshipit-source-id: 6d8772e58765485868094b92964da82ef9730b6d
2017-05-05 23:20:01 -07:00
Aaron Gao
a30a696034 do not read next datablock if upperbound is reached
Summary:
Now if we have iterate_upper_bound set, we continue read until get a key >= upper_bound. For a lot of cases that neighboring data blocks have a user key gap between them, our index key will be a user key in the middle to get a shorter size. For example, if we have blocks:
[a b c d][f g h]
Then the index key for the first block will be 'e'.
then if upper bound is any key between 'd' and 'e', for example, d1, d2, ..., d99999999999, we don't have to read the second block and also know that we have done our iteration by reaching the last key that smaller the upper bound already.

This diff can reduce RA in most cases.
Closes https://github.com/facebook/rocksdb/pull/2239

Differential Revision: D4990693

Pulled By: lightmark

fbshipit-source-id: ab30ea2e3c6edf3fddd5efed3c34fcf7739827ff
2017-05-05 23:20:01 -07:00
Maysam Yabandeh
6798d1f3be Revert "Delete filter before closing the table"
Summary:
This reverts commit 89833577a8.
Closes https://github.com/facebook/rocksdb/pull/2240

Differential Revision: D4986982

Pulled By: maysamyabandeh

fbshipit-source-id: 56c4c07b7b5b7c6fe122d5c2f2199d221c8510c0
2017-05-02 13:46:39 -07:00
Maysam Yabandeh
89833577a8 Delete filter before closing the table
Summary:
Some filters such as partitioned filter have pointers to the table for which they are created. Therefore is they are stored in the block cache, the should be forcibly erased from block cache before closing the  table, which would result into deleting the object. Otherwise the destructor will be called later when the cache is lazily erasing the object, which having the parent table no longer existent it could result into undefined behavior.

Update: there will be still cases the filter is not removed from the cache since the table has not kept a pointer to the cache handle to be able to forcibly release it later. We make sure that the filter destructor does not access the table pointer to get around such cases.
Closes https://github.com/facebook/rocksdb/pull/2207

Differential Revision: D4941591

Pulled By: maysamyabandeh

fbshipit-source-id: 56fbab2a11cf447e1aa67caa30b58d7bd7ce5bbd
2017-05-01 19:19:37 -07:00
Maysam Yabandeh
47a09b0a88 Avoid pinning when row cache is accessed
Summary:
With row cache being enabled, table cache is doing a short circuit for reading data. This path needs to be updated to take advantage of pinnable slice. In the meanwhile we disabling pinning in this path.
Closes https://github.com/facebook/rocksdb/pull/2237

Differential Revision: D4982389

Pulled By: maysamyabandeh

fbshipit-source-id: 542630d0cf23cfb1f0c397da82e7053df7966591
2017-05-01 18:12:46 -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
Tomas Kolda
04d58970cb AIX and Solaris Sparc Support
Summary:
Replacement of #2147

The change was squashed due to a lot of conflicts.
Closes https://github.com/facebook/rocksdb/pull/2194

Differential Revision: D4929799

Pulled By: siying

fbshipit-source-id: 5cd49c254737a1d5ac13f3c035f128e86524c581
2017-04-21 20:48:04 -07:00
Aaron Gao
6e8d6f429d readahead backwards from sst end
Summary:
prefetch some data from the end of the file for each compaction to reduce IO.
Closes https://github.com/facebook/rocksdb/pull/2149

Differential Revision: D4880576

Pulled By: lightmark

fbshipit-source-id: aa767cd1afc84c541837fbf1ad6c0d45b34d3932
2017-04-14 18:56:14 -07:00
Sagar Vemuri
343b59d6ee Move various string utility functions into string_util
Summary:
This is an effort to club all string related utility functions into one common place, in string_util, so that it is easier for everyone to know what string processing functions are available. Right now they seem to be spread out across multiple modules, like logging and options_helper.

Check the sub-commits for easier reviewing.
Closes https://github.com/facebook/rocksdb/pull/2094

Differential Revision: D4837730

Pulled By: sagar0

fbshipit-source-id: 344278a
2017-04-06 14:54:12 -07:00
Yi Wu
df6f5a3772 Move memtable related files into memtable directory
Summary:
Move memtable related files into memtable directory.
Closes https://github.com/facebook/rocksdb/pull/2087

Differential Revision: D4829242

Pulled By: yiwu-arbug

fbshipit-source-id: ca70ab6
2017-04-06 14:09:13 -07:00
Siying Dong
d2dce5611a Move some files under util/ to separate dirs
Summary:
Move some files under util/ to new directories env/, monitoring/ options/ and cache/
Closes https://github.com/facebook/rocksdb/pull/2090

Differential Revision: D4833681

Pulled By: siying

fbshipit-source-id: 2fd8bef
2017-04-05 19:09:16 -07:00
Aaron Gao
90cfd46458 update IterKey that can get user key and internal key explicitly
Summary:
to void future bug that caused by the mix of userkey/internalkey
Closes https://github.com/facebook/rocksdb/pull/2084

Differential Revision: D4825889

Pulled By: lightmark

fbshipit-source-id: 28411db
2017-04-04 14:24:20 -07:00
Maysam Yabandeh
e7731d119a Configure index partition size
Summary:
Allow the users to specify the target index partition size.

With this patch an index partition is cut before its estimated in-memory size goes above the configured value for metadata_block_size. The filter partitions are still cut right after an index partition is cut.
Closes https://github.com/facebook/rocksdb/pull/2041

Differential Revision: D4780216

Pulled By: maysamyabandeh

fbshipit-source-id: 95a0831
2017-03-28 12:09:12 -07:00
Maysam Yabandeh
34a70859bc Fix segmentation fault caused by #1961
Summary:
Fixes #1961 which causes a segfault when filter_policy is nullptr and both
pin_l0_filter_and_index_blocks_in_cache/cache_index_and_filter_blocks
are set.
Closes https://github.com/facebook/rocksdb/pull/2029

Differential Revision: D4764862

Pulled By: maysamyabandeh

fbshipit-source-id: 05bd695
2017-03-24 17:24:11 -07:00
Andrew Kryczka
a5c8b54340 fix a header include
Summary:
need to consistently include "rocksdb/persistent_cache.h" to fix internal build
Closes https://github.com/facebook/rocksdb/pull/2034

Differential Revision: D4768101

Pulled By: ajkr

fbshipit-source-id: 2ecb07f
2017-03-23 19:24:22 -07:00
Maysam Yabandeh
8b0097b49b Readers for partition filter
Summary:
This is the last split of this pull request: https://github.com/facebook/rocksdb/pull/1891 which includes the reader part as well as the tests.
Closes https://github.com/facebook/rocksdb/pull/1961

Differential Revision: D4672216

Pulled By: maysamyabandeh

fbshipit-source-id: 6a2b829
2017-03-22 09:24:15 -07:00
Islam AbdelRahman
e19163688b Add macros to include file name and line number during Logging
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
2017-03-15 19:39:12 -07:00
Maysam Yabandeh
11526252cc Pinnableslice (2nd attempt)
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
2017-03-13 11:54:10 -07:00
Andrew Kryczka
f2817fb7f9 avoid ASSERT_EQ(false, ...);
Summary:
lately it fails on travis due to a compiler bug (see https://github.com/google/googletest/issues/322#issuecomment-125645145). interestingly it seems to affect occurrences of `ASSERT_EQ(false, ...);` but not `ASSERT_EQ(true, ...);`.
Closes https://github.com/facebook/rocksdb/pull/1958

Differential Revision: D4680742

Pulled By: ajkr

fbshipit-source-id: 291fe41
2017-03-08 22:24:16 -08:00
Maysam Yabandeh
54b434110e Builders for partition filter
Summary:
This is the second split of this pull request: https://github.com/facebook/rocksdb/pull/1891 which includes only the builder part. The testing will be included in the third split, where the reader is also included.
Closes https://github.com/facebook/rocksdb/pull/1952

Differential Revision: D4660272

Pulled By: maysamyabandeh

fbshipit-source-id: 36b3cf0
2017-03-07 13:54:12 -08:00
Maysam Yabandeh
a2f7a514d1 Refactoring
Summary:
This is the first split of https://github.com/facebook/rocksdb/pull/1891 and will be needed for the upcoming partitioned filter patch.
Closes https://github.com/facebook/rocksdb/pull/1949

Differential Revision: D4652152

Pulled By: maysamyabandeh

fbshipit-source-id: 9801778
2017-03-03 18:24:12 -08:00
Islam AbdelRahman
f89b3893c0 Remove skip_table_builder_flush and default it to true
Summary:
This option is needed to be enabled for Direct IO
and I cannot think of a reason where we need to disable it

remove it and default it to true
Closes https://github.com/facebook/rocksdb/pull/1944

Differential Revision: D4641088

Pulled By: IslamAbdelRahman

fbshipit-source-id: d7085b9
2017-03-02 16:54:10 -08:00
Aaron Gao
e877afa08b Remove bulk loading and auto_roll_logger in rocksdb_lite
Summary:
shrink lite size
Closes https://github.com/facebook/rocksdb/pull/1929

Differential Revision: D4622059

Pulled By: siying

fbshipit-source-id: 050b796
2017-02-28 11:09:11 -08:00
Andrew Kryczka
e67232cff5 Handle failed Finish() in SST file writer
Summary:
The assertion in Abandon() fails when called after Finish() fails. Finish() already closes the builder so there's no need to call Abandon().
Closes https://github.com/facebook/rocksdb/pull/1901

Differential Revision: D4601373

Pulled By: ajkr

fbshipit-source-id: e5678be
2017-02-23 15:39:16 -08:00
Sagar Vemuri
eb912a927e Remove disableDataSync option
Summary:
Remove disableDataSync, and another similarly named disable_data_sync options.
This is being done to simplify options, and also because the performance gains of this feature can be achieved by other methods.
Closes https://github.com/facebook/rocksdb/pull/1859

Differential Revision: D4541292

Pulled By: sagar0

fbshipit-source-id: 5b3a6ca
2017-02-13 11:09:13 -08:00
Maysam Yabandeh
69d5262c81 Two-level Indexes
Summary:
Partition Index blocks and use a Partition-index as a 2nd level index.

The two-level index can be used by setting
BlockBasedTableOptions::kTwoLevelIndexSearch as the index type and
configuring BlockBasedTableOptions::index_per_partition

t15539501
Closes https://github.com/facebook/rocksdb/pull/1814

Differential Revision: D4473535

Pulled By: maysamyabandeh

fbshipit-source-id: bffb87e
2017-02-06 16:39:12 -08:00
Dmitri Smirnov
0a4cdde50a Windows thread
Summary:
introduce new methods into a public threadpool interface,
- allow submission of std::functions as they allow greater flexibility.
- add Joining methods to the implementation to join scheduled and submitted jobs with
  an option to cancel jobs that did not start executing.
- Remove ugly `#ifdefs` between pthread and std implementation, make it uniform.
- introduce pimpl for a drop in replacement of the implementation
- Introduce rocksdb::port::Thread typedef which is a replacement for std::thread.  On Posix Thread defaults as before std::thread.
- Implement WindowsThread that allocates memory in a more controllable manner than windows std::thread with a replaceable implementation.
- should be no functionality changes.
Closes https://github.com/facebook/rocksdb/pull/1823

Differential Revision: D4492902

Pulled By: siying

fbshipit-source-id: c74cb11
2017-02-06 14:54:18 -08:00
Andrew Kryczka
b797e42157 Dump compression dictionary meta-block
Summary:
make sst_dump print size/contents of the dictionary meta-block for easier debugging
Closes https://github.com/facebook/rocksdb/pull/1837

Differential Revision: D4506399

Pulled By: ajkr

fbshipit-source-id: b9bf668
2017-02-03 12:39:16 -08:00
Islam AbdelRahman
574b543f80 Rename merger.h -> merging_iterator.h
Summary:
merger.h was always a confusing name for me, simply give the file a better name
Closes https://github.com/facebook/rocksdb/pull/1836

Differential Revision: D4505357

Pulled By: IslamAbdelRahman

fbshipit-source-id: 07b28d8
2017-02-02 16:54:19 -08:00
Andrew Kryczka
3b35134e4b Avoid cache lookups for range deletion meta-block
Summary:
I added the Cache::Ref() function a couple weeks ago (#1761) to make this feature possible. Like other meta-blocks, rep_->range_del_entry holds a cache handle to pin the range deletion block in uncompressed block cache for the duration of the table reader's lifetime. We can reuse this cache handle to create an iterator over this meta-block without any cache lookup. Ref() is used to increment the cache handle's refcount in case the returned iterator outlives the table reader.
Closes https://github.com/facebook/rocksdb/pull/1801

Differential Revision: D4458782

Pulled By: ajkr

fbshipit-source-id: 2883f10
2017-01-26 11:24:13 -08:00
Maysam Yabandeh
d0ba8ec8f9 Revert "PinnableSlice"
Summary:
This reverts commit 54d94e9c2c.

The pull request was landed by mistake.
Closes https://github.com/facebook/rocksdb/pull/1755

Differential Revision: D4391678

Pulled By: maysamyabandeh

fbshipit-source-id: 36d5149
2017-01-08 14:24:12 -08:00
Maysam Yabandeh
54d94e9c2c 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:
 1. value 100 byte: 1.8%  regular, 1.2% merge values
 2. value 1k   byte: 11.5% regular, 7.5% merge values
 3. 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 difference
is a little and could be noise. More importantly it is safely
cancelled
Closes https://github.com/facebook/rocksdb/pull/1732

Differential Revision: D4374613

Pulled By: maysamyabandeh

fbshipit-source-id: a077f1a
2017-01-08 13:54:13 -08:00
Maysam Yabandeh
0712d541d1 Delegate Cleanables
Summary:
Cleanable objects will perform the registered cleanups when
they are destructed. We however rather to delay this cleaning like when
we are gathering the merge operands. Current approach is to create the
Cleanable object on heap (instead of on stack) and delay deleting it.

By allowing Cleanables to delegate their cleanups to another cleanable
object we can delay the cleaning without however the need to craete the
cleanable object on heap and keeping it around. This patch applies this
technique for the cleanups of BlockIter and shows improved performance
for some in-memory benchmarks:
+1.8% for merge worklaod, +6.4% for non-merge workload when the merge
operator is specified.
https://our.intern.facebook.com/intern/tasks?t=15168163

Non-merge benchmark:
TEST_TMPDIR=/dev/shm/v100nocomp/ ./db_bench --benchmarks=fillrandom
--num=1000000 -value_size=100 -compression_type=none

Reading random with no merge operator specified:
TEST_TMPDIR=/dev/shm/v100nocomp/ ./db_bench
--benchmarks="read
Closes https://github.com/facebook/rocksdb/pull/1711

Differential Revision: D4361163

Pulled By: maysamyabandeh

fbshipit-source-id: 9801e07
2016-12-29 15:54:19 -08:00
Islam AbdelRahman
d58ef52ba6 Allow SstFileWriter to Fadvise the file away from page cache
Summary:
Add `fadvise_trigger` option to `SstFileWriter`

If fadvise_trigger is passed with a non-zero value, SstFileWriter will invalidate the os page cache every `fadvise_trigger` bytes for the sst file
Closes https://github.com/facebook/rocksdb/pull/1731

Differential Revision: D4371246

Pulled By: IslamAbdelRahman

fbshipit-source-id: 91caff1
2016-12-29 15:09:19 -08:00
Yi Wu
ab48c165a9 Print cache options to info log
Summary:
Improve cache options logging to info log.
Also print the value of
cache_index_and_filter_blocks_with_high_priority.
Closes https://github.com/facebook/rocksdb/pull/1709

Differential Revision: D4358776

Pulled By: yiwu-arbug

fbshipit-source-id: 8f030a0
2016-12-22 14:54:19 -08:00
Yi Wu
5d1457dbbf Dump persistent cache options
Summary:
Dump persistent cache options
Closes https://github.com/facebook/rocksdb/pull/1679

Differential Revision: D4337019

Pulled By: yiwu-arbug

fbshipit-source-id: 3812f8a
2016-12-19 14:09:12 -08:00
Ding Ma
1a136c1f13 Expose file size
Summary:
add a new function to SstFileWriter that will tell the user how big is there file right now.
Closes https://github.com/facebook/rocksdb/pull/1686

Differential Revision: D4338868

Pulled By: mdyuki1016

fbshipit-source-id: c1ee16a
2016-12-16 18:39:12 -08:00
Daniel Black
5334d8b44c table/block_based_table_builder.cc: intentional fallthrough - comment to match gcc pattern
Summary:
The gcc-7 code for parsing comments (libcpp/lex.c) didn't match
the intentional fallthough in this comment.

table/block_based_table_builder.cc: In member function 'void rocksdb::BlockBasedTableBuilder::WriteRawBlock(const rocksdb::Slice&, rocksdb::CompressionType, rocksdb::BlockHandle*)':
table/block_based_table_builder.cc:754:22: error: this statement may fall through [-Werror=implicit-fallthrough=]
         assert(false);
                      ^
table/block_based_table_builder.cc:756:7: note: here
       case kCRC32c: {
       ^~~~
cc1plus: all warnings being treated as errors
Closes https://github.com/facebook/rocksdb/pull/1661

Differential Revision: D4318817

Pulled By: yiwu-arbug

fbshipit-source-id: e67d171
2016-12-13 13:54:15 -08:00
Islam AbdelRahman
67f37cf198 Allow user to specify a CF for SST files generated by SstFileWriter
Summary:
Allow user to explicitly specify that the generated file by SstFileWriter will be ingested in a specific CF.
This allow us to persist the CF id in the generated file
Closes https://github.com/facebook/rocksdb/pull/1615

Differential Revision: D4270422

Pulled By: IslamAbdelRahman

fbshipit-source-id: 7fb954e
2016-12-05 14:24:16 -08:00
Andrew Kryczka
fd43ee09da Range deletion microoptimizations
Summary:
- Made RangeDelAggregator's InternalKeyComparator member a reference-to-const so we don't need to copy-construct it. Also added InternalKeyComparator to ImmutableCFOptions so we don't need to construct one for each DBIter.
- Made MemTable::NewRangeTombstoneIterator and the table readers' NewRangeTombstoneIterator() functions return nullptr instead of NewEmptyInternalIterator to avoid the allocation. Updated callers accordingly.
Closes https://github.com/facebook/rocksdb/pull/1548

Differential Revision: D4208169

Pulled By: ajkr

fbshipit-source-id: 2fd65cf
2016-11-21 12:24:13 -08:00
Andrew Kryczka
fe349db57b Remove Arena in RangeDelAggregator
Summary:
The Arena construction/destruction introduced significant overhead to read-heavy workload just by creating empty vectors for its blocks, so avoid it in RangeDelAggregator.
Closes https://github.com/facebook/rocksdb/pull/1547

Differential Revision: D4207781

Pulled By: ajkr

fbshipit-source-id: 9d1c130
2016-11-19 14:24:12 -08:00
Siying Dong
a4eb7387b2 Allow plain table to store index on file with bloom filter disabled
Summary:
Currently plain table bloom filter is required if storing metadata on file. Remove the constraint.
Closes https://github.com/facebook/rocksdb/pull/1525

Differential Revision: D4190977

Pulled By: siying

fbshipit-source-id: be60442
2016-11-17 11:09:13 -08:00
Andrew Kryczka
327085b7b2 fix valgrind
Summary: Closes https://github.com/facebook/rocksdb/pull/1526

Differential Revision: D4191257

Pulled By: ajkr

fbshipit-source-id: d09dc76
2016-11-16 12:09:11 -08:00
Andrew Kryczka
48e8baebc0 Decouple data iterator and range deletion iterator in TableCache
Summary:
Previously we used TableCache::NewIterator() for multiple purposes (data
block iterator and range deletion iterator), and returned non-ok status in
the data block iterator. In one case where the caller only used the range
deletion block iterator (9e7cf3469b/db/version_set.cc (L965-L973)),
we didn't check/free the data block iterator containing non-ok status, which
caused a valgrind error.

So, this diff decouples creation of data block and range deletion block iterators,
and updates the callers accordingly. Both functions can return non-ok status
in an InternalIterator. Since the non-ok status is returned in an iterator that the
callers will definitely use, it should be more usable/less error-prone.
Closes https://github.com/facebook/rocksdb/pull/1513

Differential Revision: D4181423

Pulled By: ajkr

fbshipit-source-id: 835b8f5
2016-11-15 17:24:28 -08:00
Islam AbdelRahman
5ed650857d Fix SstFileWriter destructor
Summary:
If user did not call SstFileWriter::Finish() or called Finish() but it failed.
We need to abandon the builder, to avoid destructing it while it's open
Closes https://github.com/facebook/rocksdb/pull/1502

Differential Revision: D4171660

Pulled By: IslamAbdelRahman

fbshipit-source-id: ab6f434
2016-11-12 20:11:19 -08:00
Andrew Kryczka
307a4e80c8 sst_dump support for range deletion
Summary:
Change DumpTable() so we can see the range deletion meta-block.
Closes https://github.com/facebook/rocksdb/pull/1505

Differential Revision: D4172227

Pulled By: ajkr

fbshipit-source-id: ae35665
2016-11-12 09:39:23 -08:00
Andrew Kryczka
815f54afad Insert range deletion meta-block into block cache
Summary:
This handles two issues: (1) range deletion iterator sometimes outlives
the table reader that created it, in which case the block must not be destroyed
during table reader destruction; and (2) we prefer to read these range tombstone
meta-blocks from file fewer times.

- Extracted cache-populating logic from NewDataBlockIterator() into a separate function: MaybeLoadDataBlockToCache()
- Use MaybeLoadDataBlockToCache() to load range deletion meta-block and pin it through the reader's lifetime. This code reuse works since range deletion meta-block has same format as data blocks.
- Use NewDataBlockIterator() to create range deletion iterators, which uses block cache if enabled, otherwise reads the block from file. Either way, the underlying block won't disappear until after the iterator is destroyed.
Closes https://github.com/facebook/rocksdb/pull/1459

Differential Revision: D4123175

Pulled By: ajkr

fbshipit-source-id: 8f64281
2016-11-05 09:24:26 -07:00
Islam AbdelRahman
5c5d01ae74 Fix wrong comment (Maximum supported block size)
Summary:
We can support SST files >2GB but we don't support blocks >2GB
Closes https://github.com/facebook/rocksdb/pull/1465

Differential Revision: D4132140

Pulled By: yiwu-arbug

fbshipit-source-id: 63bf12d
2016-11-04 11:24:14 -07:00
Andrew Kryczka
f998c9790f DeleteRange Get support
Summary:
During Get()/MultiGet(), build up a RangeDelAggregator with range
tombstones as we search through live memtable, immutable memtables, and
SST files. This aggregator is then used by memtable.cc's SaveValue() and
GetContext::SaveValue() to check whether keys are covered.

added tests for Get on memtables/files; end-to-end tests mainly in https://reviews.facebook.net/D64761
Closes https://github.com/facebook/rocksdb/pull/1456

Differential Revision: D4111271

Pulled By: ajkr

fbshipit-source-id: 6e388d4
2016-11-03 18:54:20 -07:00
Andrew Kryczka
40a2e406f8 DeleteRange flush support
Summary:
Changed BuildTable() (used for flush) to (1) add range
tombstones to the aggregator, which is used by CompactionIterator to
determine which keys can be removed; and (2) add aggregator's range
tombstones to the table that is output for the flush.
Closes https://github.com/facebook/rocksdb/pull/1438

Differential Revision: D4100025

Pulled By: ajkr

fbshipit-source-id: cb01a70
2016-10-31 20:54:18 -07:00
Willem Jan Withagen
0aab5e55f0 FreeBSD: malloc_usable_size is in <malloc_np.h> (#1428)
Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>
2016-10-28 10:44:52 -07:00
Aaron Gao
9de2f75216 revert Prev() in MergingIterator to use previous code in non-prefix-seek mode
Summary: Siying suggested to keep old code for normal mode prev() for safety

Test Plan: make check -j64

Reviewers: yiwu, andrewkr, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D65439
2016-10-24 13:13:01 -07:00
Aaron Gao
59a7c0337b Change ioptions to store user_comparator, fix bug
Summary:
change ioptions.comparator to user_comparator instread of internal_comparator.
Also change Comparator* to InternalKeyComparator* to make its type explicitly.

Test Plan: make all check -j64

Reviewers: andrewkr, sdong, yiwu

Reviewed By: yiwu

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D65121
2016-10-21 11:31:42 -07:00
Andrew Kryczka
a0ba0aa877 Fix uninitialized variable gcc error for MyRocks
Summary: make sure seq_ is properly initialized even if ParseInternalKey() fails.

Test Plan: run myrocks release tests

Reviewers: lightmark, mung, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D65199
2016-10-19 10:59:46 -07:00
Islam AbdelRahman
b88f8e87c5 Support SST files with Global sequence numbers [reland]
Summary:
reland https://reviews.facebook.net/D62523

- Update SstFileWriter to include a property for a global sequence number in the SST file `rocksdb.external_sst_file.global_seqno`
- Update TableProperties to be aware of the offset of each property in the file
- Update BlockBasedTableReader and Block to be able to honor the sequence number in `rocksdb.external_sst_file.global_seqno` property and use it to overwrite all sequence number in the file

Something worth mentioning is that we don't update the seqno in the index block since and when doing a binary search, the reason for that is that it's guaranteed that SST files with global seqno will have only one user_key and each key will have seqno=0 encoded in it, This mean that this key is greater than any other key with seqno> 0. That mean that we can actually keep the current logic for these blocks

Test Plan: unit tests

Reviewers: sdong, yhchiang

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D65211
2016-10-18 16:59:37 -07:00
Andrew Kryczka
6fbe96baf8 Compaction Support for Range Deletion
Summary:
This diff introduces RangeDelAggregator, which takes ownership of iterators
provided to it via AddTombstones(). The tombstones are organized in a two-level
map (snapshot stripe -> begin key -> tombstone). Tombstone creation avoids data
copy by holding Slices returned by the iterator, which remain valid thanks to pinning.

For compaction, we create a hierarchical range tombstone iterator with structure
matching the iterator over compaction input data. An aggregator based on that
iterator is used by CompactionIterator to determine which keys are covered by
range tombstones. In case of merge operand, the same aggregator is used by
MergeHelper. Upon finishing each file in the compaction, relevant range tombstones
are added to the output file's range tombstone metablock and file boundaries are
updated accordingly.

To check whether a key is covered by range tombstone, RangeDelAggregator::ShouldDelete()
considers tombstones in the key's snapshot stripe. When this function is used outside of
compaction, it also checks newer stripes, which can contain covering tombstones. Currently
the intra-stripe check involves a linear scan; however, in the future we plan to collapse ranges
within a stripe such that binary search can be used.

RangeDelAggregator::AddToBuilder() adds all range tombstones in the table's key-range
to a new table's range tombstone meta-block. Since range tombstones may fall in the gap
between files, we may need to extend some files' key-ranges. The strategy is (1) first file
extends as far left as possible and other files do not extend left, (2) all files extend right
until either the start of the next file or the end of the last range tombstone in the gap,
whichever comes first.

One other notable change is adding release/move semantics to ScopedArenaIterator
such that it can be used to transfer ownership of an arena-allocated iterator, similar to
how unique_ptr is used for malloc'd data.

Depends on D61473

Test Plan: compaction_iterator_test, mock_table, end-to-end tests in D63927

Reviewers: sdong, IslamAbdelRahman, wanning, yhchiang, lightmark

Reviewed By: lightmark

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D62205
2016-10-18 12:04:56 -07:00
Aaron Gao
21e8daced5 fix assertion failure in Prev()
Summary:
fix assertion failure in db_stress.
It happens because of prefix seek key is larger than merge iterator key when they have the same user key

Test Plan: ./db_stress --max_background_compactions=1 --max_write_buffer_number=3 --sync=0 --reopen=20 --write_buffer_size=33554432 --delpercent=5 --log2_keys_per_lock=10 --block_size=16384 --allow_concurrent_memtable_write=0 --test_batches_snapshots=0 --max_bytes_for_level_base=67108864 --progress_reports=0 --mmap_read=0 --writepercent=35 --disable_data_sync=0 --readpercent=50 --subcompactions=4 --ops_per_thread=20000000 --memtablerep=skip_list --prefix_size=0 --target_file_size_multiplier=1 --column_families=1 --threads=32 --disable_wal=0 --open_files=500000 --destroy_db_initially=0 --target_file_size_base=16777216 --nooverwritepercent=1 --iterpercent=10 --max_key=100000000 --prefixpercent=0 --use_clock_cache=false --kill_random_test=888887 --cache_size=1048576 --verify_checksum=1

Reviewers: sdong, andrewkr, yiwu, yhchiang

Reviewed By: yhchiang

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D65025
2016-10-13 17:36:48 -07:00
Dmitri Smirnov
e489270980 Fix scoped arena iterator (#1387) 2016-10-12 11:16:16 -07:00
Aaron Gao
447f17127c new Prev() prefix support using SeekForPrev()
Summary:
1) The previous solution for Prev() prefix support is not clean.
Since I add api SeekForPrev(), now the Prev() can be symmetric to Next().
and we do not need SeekToLast() to be called in Prev() any more.

Also, Next() will Seek(prefix_seek_key_) to solve the problem of possible inconsistency between db_iter and merge_iter when
there is merge_operator. And prefix_seek_key is only refreshed when change direction to forward.

2) This diff also solves the bug of Iterator::SeekToLast() with iterate_upper_bound_ with prefix extractor.

add test cases for the above two cases.

There are some tests for the SeekToLast() in Prev(), I will clean them later.

Test Plan: make all check

Reviewers: IslamAbdelRahman, andrewkr, yiwu, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D63933
2016-10-11 13:54:26 -07:00
Yi Wu
991b585ee0 More block cache tickers
Summary: Adding several missing block cache tickers.

Test Plan:
  make all check

Reviewers: IslamAbdelRahman, yhchiang, lightmark

Reviewed By: lightmark

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D64881
2016-10-11 11:59:05 -07:00
Islam AbdelRahman
d062328977 Revert "Support SST files with Global sequence numbers"
This reverts commit ab01da5437.
2016-10-07 14:05:12 -07:00
Islam AbdelRahman
ab01da5437 Support SST files with Global sequence numbers
Summary:
- Update SstFileWriter to include a property for a global sequence number in the SST file `rocksdb.external_sst_file.global_seqno`
- Update TableProperties to be aware of the offset of each property in the file
- Update BlockBasedTableReader and Block to be able to honor the sequence number in `rocksdb.external_sst_file.global_seqno` property and use it to overwrite all sequence number in the file

Something worth mentioning is that we don't update the seqno in the index block since and when doing a binary search, the reason for that is that it's guaranteed that SST files with global seqno will have only one user_key and each key will have seqno=0 encoded in it, This mean that this key is greater than any other key with seqno> 0. That mean that we can actually keep the current logic for these blocks

Test Plan: unit tests

Reviewers: andrewkr, yhchiang, yiwu, sdong

Reviewed By: sdong

Subscribers: hcz, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D62523
2016-10-03 16:12:39 -07:00
Andrew Kryczka
6009c473c7 Store range tombstones in memtable
Summary:
- Store range tombstones in a separate MemTableRep instantiated with ColumnFamilyOptions::memtable_factory
- MemTable::NewRangeTombstoneIterator() returns a MemTableIterator over the separate MemTableRep
- Part of the read path is not implemented yet (i.e., MemTable::Get())

Test Plan: see unit tests

Reviewers: wanning

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D62217
2016-09-30 09:06:43 -07:00
Aaron Gao
f517d9dd09 Add SeekForPrev() to Iterator
Summary:
Add new Iterator API, `SeekForPrev`: find the last key that <= target key
support prefix_extractor
support prefix_same_as_start
support upper_bound
not supported in iterators without Prev()

Also add tests in db_iter_test and db_iterator_test

Pass all tests
Cheers!

Test Plan: make all check -j64

Reviewers: andrewkr, yiwu, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D64149
2016-09-27 18:20:57 -07:00
panfengfeng
7afbb7420b solve the problem of table_factory_to_write_=nullptr (#1342) 2016-09-20 10:11:51 -07:00
rockeet
4c3f4496b5 Add TableBuilderOptions::level and relevant changes (#1335) 2016-09-17 22:30:43 -07:00
sdong
3edb9461b7 Avoid hard-coded sleep in EnvPosixTestWithParam.TwoPools
Summary: EnvPosixTestWithParam.TwoPools relies on explicit sleeping, so it sometimes fail. Fix it.

Test Plan: Run tests with high parallelism many times and make sure the test passes.

Reviewers: yiwu, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D63417
2016-09-16 17:45:12 -07:00
Yi Wu
81747f1be6 Refactor MutableCFOptions
Summary:
* Change constructor of MutableCFOptions to depends only on ColumnFamilyOptions.
* Move `max_subcompactions`, `compaction_options_fifo` and `compaction_pri` to ImmutableCFOptions to make it clear that they are immutable.

Test Plan: existing unit tests.

Reviewers: yhchiang, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D63945
2016-09-13 21:11:59 -07:00
Islam AbdelRahman
1cca091298 Temporarily revert Prev() prefix support
Summary:
Temporarily revert commits for supporting prefix Prev() to unblock MyRocks and RocksDB release

These are the commits reverted

  - 6a14d55bd9
  - b18f9c9eac
  - db74b1a219
  - 2482d5fb45

Test Plan: make check -j64

Reviewers: sdong, lightmark

Reviewed By: lightmark

Subscribers: andrewkr, dhruba, yoshinorim

Differential Revision: https://reviews.facebook.net/D63789
2016-09-08 14:45:32 -07:00
sdong
607628d349 Support ZSTD with finalized format
Summary:
ZSTD 1.0.0 is coming. We can finally add a support of ZSTD without worrying about compatibility.
Still keep ZSTDNotFinal for compatibility reason.

Test Plan: Run all tests. Run db_bench with ZSTD version with RocksDB built with ZSTD 1.0 and older.

Reviewers: andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: cyan, igor, IslamAbdelRahman, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D63141
2016-09-06 12:22:16 -07:00
Injun Song
ce1be2ce37 Fix build error on Windows (AppVeyor) (#1315)
Add 'cf_options' to source list and db_imple.cc

fix casting
2016-09-06 08:41:43 -07:00
sdong
f7669b40ba Fix Windows Build
Summary: Fix two Windows build problems.

Test Plan: Build on Windows and run all Linux tests.

Reviewers: IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D63189
2016-09-02 17:10:28 -07:00
Yi Wu
a88677d2cf Remove ImmutableCFOptions from public API
Summary: There's no reference to ImmutableCFOptions elsewhere in /include/rocksdb. ImmutableCFOptions was introduced in this commit (5665e5e285) but later its reference in /include/rocksdb/table.h is removed.

Test Plan:
  make all check

Reviewers: IslamAbdelRahman, sdong, yhchiang

Reviewed By: yhchiang

Subscribers: yhchiang, andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D63177
2016-09-02 14:16:31 -07:00
Aaron Gao
4590b53a4b add stats to Cache::LookUp()
Summary: basically for SimCache stats. I find most times it is hard to pass Statistics* to SimCache constructor.

Test Plan: make all check

Reviewers: andrewkr, sdong, yiwu

Reviewed By: yiwu

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D62193
2016-09-01 13:50:39 -07:00
Islam AbdelRahman
8ce1b8440a Fix Travis on Mac
Summary: not sure why travis complain about this line, works fine on my mac

Test Plan: run on my mac

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D63045
2016-08-31 15:10:12 -07:00
Aaron Gao
db74b1a219 fix bug in merge_iterator when data race happens
Summary:
core dump when run
`./db_stress --max_background_compactions=1 --max_write_buffer_number=3 --sync=0 --reopen=20 --write_buffer_size=33554432 --delpercent=5 --log2_keys_per_lock=10 --block_size=16384 --allow_concurrent_memtable_write=1 --test_batches_snapshots=0 --max_bytes_for_level_base=67108864 --progress_reports=0 --mmap_read=1 --kill_prefix_blacklist=WritableFileWriter::Append,WritableFileWriter::WriteBuffered --writepercent=35 --disable_data_sync=0 --readpercent=50 --subcompactions=3 --ops_per_thread=20000000 --memtablerep=skip_list --prefix_size=0 --target_file_size_multiplier=1 --column_families=1 --db=/dev/shm/rocksdb/rocksdb_crashtest_whitebox --threads=32 --disable_wal=0 --open_files=500000 --destroy_db_initially=0 --target_file_size_base=16777216 --nooverwritepercent=1 --iterpercent=10 --max_key=100000000 --prefixpercent=0 --use_clock_cache=false --kill_random_test=189 --cache_size=1048576 --verify_checksum=1`
Actually the relevant flag is `--threads`, data race when --thread > 1 cause problem.
It is possible that multiple
threads read/write memtable simultaneously. After one thread
calls Prev(), another thread may insert a new key just between
the current key and the key next, which may cause the
assert(current_ == CurrentForward()) failure when the first
thread calls Next() again if in prefix seek mode

Test Plan: rerun db_stress with >1 thread / make all check -j64

Reviewers: sdong, andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D62979
2016-08-30 22:19:42 -07:00
Aaron Gao
b18f9c9eac add nullptr check to internal_prefix_transform
Summary: patch for D62361

Test Plan: make all check

Reviewers: sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D62883
2016-08-30 13:48:31 -07:00
Aaron Gao
2482d5fb45 support Prev() in prefix seek mode
Summary: As title, make sure Prev() works as expected with Next() when the current iter->key() in the range of the same prefix in prefix seek mode

Test Plan: make all check -j64 (add prefix_test with PrefixSeekModePrev test case)

Reviewers: andrewkr, sdong, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: yoshinorim, andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D61419
2016-08-29 20:55:39 -07:00
Islam AbdelRahman
b49b92cf28 Introduce Read amplification bitmap (read amp statistics)
Summary:
Add ReadOptions::read_amp_bytes_per_bit option which allow us to create a bitmap for every data block we read
the bitmap will contain (block_size / read_amp_bytes_per_bit) bits.

We will use this bitmap to mark which bytes have been used of the block so we can calculate the read amplification

Test Plan: added new tests

Reviewers: andrewkr, yhchiang, sdong

Reviewed By: sdong

Subscribers: yiwu, leveldb, march, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D58707
2016-08-26 18:55:58 -07:00
Aaron Gao
c7004840d2 store prefix_extractor_name in table
Summary:
Make sure prefix extractor name is stored in SST files and if DB is opened with a prefix extractor of a different name, prefix bloom is skipped when read the file.
Also add unit tests for that.

Test Plan:
before change:
```
Note: Google Test filter = BlockBasedTableTest.SkipPrefixBloomFilter
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from BlockBasedTableTest
[ RUN      ] BlockBasedTableTest.SkipPrefixBloomFilter
table/table_test.cc:1421: Failure
Value of: db_iter->Valid()
Actual: false
Expected: true
[  FAILED  ] BlockBasedTableTest.SkipPrefixBloomFilter (1 ms)
[----------] 1 test from BlockBasedTableTest (1 ms total)

[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (1 ms total)
[  PASSED  ] 0 tests.
[  FAILED  ] 1 test, listed below:
[  FAILED  ] BlockBasedTableTest.SkipPrefixBloomFilter

1 FAILED TEST
```
after:
```
Note: Google Test filter = BlockBasedTableTest.SkipPrefixBloomFilter
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from BlockBasedTableTest
[ RUN      ] BlockBasedTableTest.SkipPrefixBloomFilter
[       OK ] BlockBasedTableTest.SkipPrefixBloomFilter (0 ms)
[----------] 1 test from BlockBasedTableTest (0 ms total)

[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (0 ms total)
[  PASSED  ] 1 test.
```

Reviewers: sdong, andrewkr, yiwu, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D61215
2016-08-26 11:46:32 -07:00
Aaron Gao
cec2c6436b fix data race in NewIndexIterator() in block_based_table_reader.cc
Summary: fixed data race described in https://github.com/facebook/rocksdb/issues/1267 and add regression test

Test Plan:
./table_test --gtest_filter=BlockBasedTableTest.NewIndexIteratorLeak
make all check -j64
core dump before fix. ok after fix.

Reviewers: andrewkr, sdong

Reviewed By: sdong

Subscribers: igor, andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D62361
2016-08-23 18:20:41 -07:00
Yi Wu
4a16c32ece Option to cache index/filter blocks with priority
Summary:
Add option to block based table to insert index/filter blocks to block cache with priority. Combined with LRUCache with high_pri_pool_ratio, we can reserved space for index/filter blocks, make them less likely to be evicted.

Depends on D61977.

Test Plan: See unit test.

Reviewers: lightmark, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, march, leveldb

Differential Revision: https://reviews.facebook.net/D62241
2016-08-23 13:44:13 -07:00
Andrew Kryczka
ecf9003860 Fix bug in printing values for block-based table
Summary: value is not an InternalKey, we do not need to decode it

Test Plan:
setup:

  $ ldb put --create_if_missing=true k v
  $ ldb put --db=./tmp --create_if_missing k v
  $ ldb compact --db=./tmp

before:

  $ sst_dump --command=raw --file=./tmp/000004.sst
  ...
  terminate called after throwing an instance of 'std::length_error'

after:

  $ ./sst_dump --command=raw --file=./tmp/000004.sst
  $ cat tmp/000004_dump.txt
  ...
  ASCII  k : v
  ...

Reviewers: sdong, yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D62301
2016-08-22 10:27:50 -07:00
Islam AbdelRahman
6a17b07ca8 Add TablePropertiesCollector support in SstFileWriter
Summary: Update SstFileWriter to use user TablePropertiesCollectors that are passed in Options

Test Plan: unittests

Reviewers: sdong

Reviewed By: sdong

Subscribers: jkedgar, andrewkr, hermanlee4, dhruba, yoshinorim

Differential Revision: https://reviews.facebook.net/D62253
2016-08-19 16:17:56 -07:00
Wanning Jiang
78837f5d61 TableBuilder / TableReader support for range deletion
Summary: 1. Range Deletion Tombstone structure 2. Modify Add() in table_builder to make it usable for adding range del tombstones 3. Expose NewTombstoneIterator() API in table_reader

Test Plan: table_test.cc (now BlockBasedTableBuilder::Add() only accepts InternalKey. I make table_test only pass InternalKey to BlockBasedTableBuidler. Also test writing/reading range deletion tombstones in table_test )

Reviewers: sdong, IslamAbdelRahman, lightmark, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D61473
2016-08-19 15:10:31 -07:00
Philipp Unterbrunner
deda159b55 Added min/max/avg data block size output to sst_dump
Summary: Added min/max/avg data block size output to sst_dump. Output was added to the end of BlockBasedTable::DumpDataBlocks, so it appears after the data block details, at the very end of the dump file.

Test Plan:
```
./db_bench --benchmarks=fillrandom
./sst_dump --file=/tmp/rocksdbtest-xyz/dbbench/000007.sst --command=raw
tail -n 6 /tmp/rocksdbtest-xyz/dbbench/000007_dump.txt
```
```
Data Block Summary:
--------------------------------------
  # data blocks: 11336
  min data block size: 903
  max data block size: 2268
  avg data block size: 2245.363356
```

Reviewers: IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D61815
2016-08-12 16:34:11 -07:00
Islam AbdelRahman
b693ba68b5 Minor PinnedIteratorsManager Refactoring
Summary:
This diff include these simple change
- Rename ReleasePinnedIterators to ReleasePinnedData
- Rename PinIteratorIfNeeded to PinIterator
- Use std::vector directly in PinnedIteratorsManager instead of std::unique_ptr<std::vector>
- Generalize PinnedIteratorsManager by adding PinPtr which can pin any pointer

Test Plan: existing tests

Reviewers: sdong, yiwu, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D61305
2016-08-11 11:54:17 -07:00
omegaga
d51dc96a79 Experiments on column-aware encodings
Summary:
Experiments on column-aware encodings. Supported features: 1) extract data blocks from SST file and encode with specified encodings; 2) Decode encoded data back into row format; 3) Directly extract data blocks and write in row format (without prefix encoding); 4) Get column distribution statistics for column format; 5) Dump data blocks separated by columns in human-readable format.

There is still on-going work on this diff. More refactoring is necessary.

Test Plan: Wrote tests in `column_aware_encoding_test.cc`. More tests should be added.

Reviewers: sdong

Reviewed By: sdong

Subscribers: arahut, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D60027
2016-08-01 14:50:19 -07:00
omegaga
e70020e4f6 Only cache level 0 indexes and filter when opening table reader
Summary: In T8216281 we decided to disable prefetching the index and filter during opening table handlers during startup (max_open_files = -1).

Test Plan: Rely on `IndexAndFilterBlocksOfNewTableAddedToCache` to guarantee L0 indexes and filters are still cached and change `PinL0IndexAndFilterBlocksTest` to make sure other levels are not cached (maybe add one more test to test we don't cache other levels?)

Reviewers: sdong, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59913
2016-07-20 11:23:31 -07:00
Islam AbdelRahman
68a8e6b8fa Introduce FullMergeV2 (eliminate memcpy from merge operators)
Summary:
This diff update the code to pin the merge operator operands while the merge operation is done, so that we can eliminate the memcpy cost, to do that we need a new public API for FullMerge that replace the std::deque<std::string> with std::vector<Slice>

This diff is stacked on top of D56493 and D56511

In this diff we
- Update FullMergeV2 arguments to be encapsulated in MergeOperationInput and MergeOperationOutput which will make it easier to add new arguments in the future
- Replace std::deque<std::string> with std::vector<Slice> to pass operands
- Replace MergeContext std::deque with std::vector (based on a simple benchmark I ran https://gist.github.com/IslamAbdelRahman/78fc86c9ab9f52b1df791e58943fb187)
- Allow FullMergeV2 output to be an existing operand

```
[Everything in Memtable | 10K operands | 10 KB each | 1 operand per key]

DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="mergerandom,readseq,readseq,readseq,readseq,readseq" --merge_operator="max" --merge_keys=10000 --num=10000 --disable_auto_compactions --value_size=10240 --write_buffer_size=1000000000

[FullMergeV2]
readseq      :       0.607 micros/op 1648235 ops/sec; 16121.2 MB/s
readseq      :       0.478 micros/op 2091546 ops/sec; 20457.2 MB/s
readseq      :       0.252 micros/op 3972081 ops/sec; 38850.5 MB/s
readseq      :       0.237 micros/op 4218328 ops/sec; 41259.0 MB/s
readseq      :       0.247 micros/op 4043927 ops/sec; 39553.2 MB/s

[master]
readseq      :       3.935 micros/op 254140 ops/sec; 2485.7 MB/s
readseq      :       3.722 micros/op 268657 ops/sec; 2627.7 MB/s
readseq      :       3.149 micros/op 317605 ops/sec; 3106.5 MB/s
readseq      :       3.125 micros/op 320024 ops/sec; 3130.1 MB/s
readseq      :       4.075 micros/op 245374 ops/sec; 2400.0 MB/s
```

```
[Everything in Memtable | 10K operands | 10 KB each | 10 operand per key]

DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="mergerandom,readseq,readseq,readseq,readseq,readseq" --merge_operator="max" --merge_keys=1000 --num=10000 --disable_auto_compactions --value_size=10240 --write_buffer_size=1000000000

[FullMergeV2]
readseq      :       3.472 micros/op 288018 ops/sec; 2817.1 MB/s
readseq      :       2.304 micros/op 434027 ops/sec; 4245.2 MB/s
readseq      :       1.163 micros/op 859845 ops/sec; 8410.0 MB/s
readseq      :       1.192 micros/op 838926 ops/sec; 8205.4 MB/s
readseq      :       1.250 micros/op 800000 ops/sec; 7824.7 MB/s

[master]
readseq      :      24.025 micros/op 41623 ops/sec;  407.1 MB/s
readseq      :      18.489 micros/op 54086 ops/sec;  529.0 MB/s
readseq      :      18.693 micros/op 53495 ops/sec;  523.2 MB/s
readseq      :      23.621 micros/op 42335 ops/sec;  414.1 MB/s
readseq      :      18.775 micros/op 53262 ops/sec;  521.0 MB/s

```

```
[Everything in Block cache | 10K operands | 10 KB each | 1 operand per key]

[FullMergeV2]
$ DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readseq,readseq,readseq,readseq,readseq" --merge_operator="max" --num=100000 --db="/dev/shm/merge-random-10K-10KB" --cache_size=1000000000 --use_existing_db --disable_auto_compactions
readseq      :      14.741 micros/op 67837 ops/sec;  663.5 MB/s
readseq      :       1.029 micros/op 971446 ops/sec; 9501.6 MB/s
readseq      :       0.974 micros/op 1026229 ops/sec; 10037.4 MB/s
readseq      :       0.965 micros/op 1036080 ops/sec; 10133.8 MB/s
readseq      :       0.943 micros/op 1060657 ops/sec; 10374.2 MB/s

[master]
readseq      :      16.735 micros/op 59755 ops/sec;  584.5 MB/s
readseq      :       3.029 micros/op 330151 ops/sec; 3229.2 MB/s
readseq      :       3.136 micros/op 318883 ops/sec; 3119.0 MB/s
readseq      :       3.065 micros/op 326245 ops/sec; 3191.0 MB/s
readseq      :       3.014 micros/op 331813 ops/sec; 3245.4 MB/s
```

```
[Everything in Block cache | 10K operands | 10 KB each | 10 operand per key]

DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readseq,readseq,readseq,readseq,readseq" --merge_operator="max" --num=100000 --db="/dev/shm/merge-random-10-operands-10K-10KB" --cache_size=1000000000 --use_existing_db --disable_auto_compactions

[FullMergeV2]
readseq      :      24.325 micros/op 41109 ops/sec;  402.1 MB/s
readseq      :       1.470 micros/op 680272 ops/sec; 6653.7 MB/s
readseq      :       1.231 micros/op 812347 ops/sec; 7945.5 MB/s
readseq      :       1.091 micros/op 916590 ops/sec; 8965.1 MB/s
readseq      :       1.109 micros/op 901713 ops/sec; 8819.6 MB/s

[master]
readseq      :      27.257 micros/op 36687 ops/sec;  358.8 MB/s
readseq      :       4.443 micros/op 225073 ops/sec; 2201.4 MB/s
readseq      :       5.830 micros/op 171526 ops/sec; 1677.7 MB/s
readseq      :       4.173 micros/op 239635 ops/sec; 2343.8 MB/s
readseq      :       4.150 micros/op 240963 ops/sec; 2356.8 MB/s
```

Test Plan: COMPILE_WITH_ASAN=1 make check -j64

Reviewers: yhchiang, andrewkr, sdong

Reviewed By: sdong

Subscribers: lovro, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D57075
2016-07-20 09:49:03 -07:00
John Alexander
9430333f84 New Statistics to track Compression/Decompression (#1197)
* Added new statistics and refactored to allow ioptions to be passed around as required to access environment and statistics pointers (and, as a convenient side effect, info_log pointer).

* Prevent incrementing compression counter when compression is turned off in options.

* Prevent incrementing compression counter when compression is turned off in options.

* Added two more supported compression types to test code in db_test.cc

* Prevent incrementing compression counter when compression is turned off in options.

* Added new StatsLevel that excludes compression timing.

* Fixed casting error in coding.h

* Fixed CompressionStatsTest for new StatsLevel.

* Removed unused variable that was breaking the Linux build
2016-07-19 09:44:03 -07:00
Jay Edgar
efd013d6d8 Miscellaneous performance improvements
Summary:
I was investigating performance issues in the SstFileWriter and found all of the following:

- The SstFileWriter::Add() function created a local InternalKey every time it was called generating a allocation and free each time.  Changed to have an InternalKey member variable that can be reset with the new InternalKey::Set() function.
- In SstFileWriter::Add() the smallest_key and largest_key values were assigned the result of a ToString() call, but it is simpler to just assign them directly from the user's key.
- The Slice class had no move constructor so each time one was returned from a function a new one had to be allocated, the old data copied to the new, and the old one was freed.  I added the move constructor which also required a copy constructor and assignment operator.
- The BlockBuilder::CurrentSizeEstimate() function calculates the current estimate size, but was being called 2 or 3 times for each key added.  I changed the class to maintain a running estimate (equal to the original calculation) so that the function can return an already calculated value.
- The code in BlockBuilder::Add() that calculated the shared bytes between the last key and the new key duplicated what Slice::difference_offset does, so I replaced it with the standard function.
- BlockBuilder::Add() had code to copy just the changed portion into the last key value (and asserted that it now matched the new key).  It is more efficient just to copy the whole new key over.
- Moved this same code up into the 'if (use_delta_encoding_)' since the last key value is only needed when delta encoding is on.
- FlushBlockBySizePolicy::BlockAlmostFull calculated a standard deviation value each time it was called, but this information would only change if block_size of block_size_deviation changed, so I created a member variable to hold the value to avoid the calculation each time.
- Each PutVarint??() function has a buffer and calls std::string::append().  Two or three calls in a row could share a buffer and a single call to std::string::append().

Some of these will be helpful outside of the SstFileWriter.  I'm not 100% the addition of the move constructor is appropriate as I wonder why this wasn't done before - maybe because of compiler compatibility?  I tried it on gcc 4.8 and 4.9.

Test Plan: The changes should not affect the results so the existing tests should all still work and no new tests were added.  The value of the changes was seen by manually testing the SstFileWriter class through MyRocks and adding timing code to identify problem areas.

Reviewers: sdong, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59607
2016-07-12 14:15:32 -07:00
Yi Wu
296545a2c7 Fix clang analyzer errors
Summary:
Fixing erros reported by clang static analyzer.
* Removing some unused variables.
* Adding assertions to fix false positives reported by clang analyzer.
* Adding `__clang_analyzer__` macro to suppress false positive warnings.

Test Plan:
    USE_CLANG=1 OPT=-g make analyze -j64

Reviewers: andrewkr, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D60549
2016-07-08 17:50:51 -07:00
sdong
32df9733d1 Add options.write_buffer_manager: control total memtable size across DB instances
Summary: Add option write_buffer_manager to help users control total memory spent on memtables across multiple DB instances.

Test Plan: Add a new unit test.

Reviewers: yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: adela, benj, sumeet, muthu, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59925
2016-07-05 18:11:25 -07:00
Islam AbdelRahman
88a2776db5 Update SstFileWriter to use bottommost_compression if avaliable
Summary: SstFileWriter ignore Options::bottommost_compression, update it to use bottommost_compression if available

Test Plan:
make check -j64
verified used compression using ./sst_dump

Reviewers: sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, yoshinorim

Differential Revision: https://reviews.facebook.net/D59841
2016-06-20 11:26:25 -07:00
Islam AbdelRahman
8366e10ffc Fix clang build
Summary: Fix clang build

Test Plan: USE_CLANG=1 make check -j64

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D59667
2016-06-15 00:24:33 -07:00
Islam AbdelRahman
812dbfb483 Optimize BlockIter::Prev() by caching decoded entries
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
2016-06-14 12:27:46 -07:00