Summary:
Added an option for readrandom benchmark to run with tailing iterator instead of Get. Benefit of tailing iterator is that it doesn't require locking DB mutex on access.
I also have some results when running on my machine. The results highly depend on number of cache shards. With our current benchmark setting of 4 table cache shards and 6 block cache shards, I don't see much improvements of using tailing iterator. In that case, we're probably seeing cache mutex contention.
Here are the results for different number of shards
cache shards tailing iterator get
6 1.38M 1.16M
10 1.58M 1.15M
As soon as we get rid of cache mutex contention, we're seeing big improvements in using tailing iterator vs. ordinary get.
Test Plan: ran regression test
Reviewers: dhruba, haobo, ljin, kailiu, sding
Reviewed By: haobo
CC: tnovak
Differential Revision: https://reviews.facebook.net/D15867
Summary:
Use super_version insider NewIterator to avoid Ref() each component
separately under mutex
The new added bench shows NewIterator QPS increases from 515K to 719K
No meaningful improvement for multiget I guess due to its relatively small
cost comparing to 90 keys fetch in the test.
Test Plan: unit test and db_bench
Reviewers: igor, sdong
Reviewed By: igor
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D15609
Summary: By removing some includes form options.h and reply on forward declaration, we can more easily reason the dependencies.
Test Plan: make all check
Reviewers: kailiu, haobo, igor, dhruba
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15411
Summary:
It looks like we might have some trouble when building the new release with 4.8, since fbcode is using glibc2.17-fb by default and we are using glibc2.17. It was reported by Benjamin Renard in our internal group.
This diff moves our fbcode build to use glibc2.17-fb by default. I got some linker errors when compiling, complaining that `google::SetUsageMessage()` was undefined. After deleting all offending lines, the compile was successful and everything works.
Test Plan:
Compiled
Ran ./db_bench ./db_stress ./db_repl_stress
Reviewers: kailiu
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15405
Summary: I'm separating code-cleanup part of https://reviews.facebook.net/D14517. This will make D14517 easier to understand and this diff easier to review.
Test Plan: make check
Reviewers: haobo, kailiu, sdong, dhruba, tnovak
Reviewed By: tnovak
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15099
Summary:
Added an option (max_successive_merges) that can be used to specify the
maximum number of successive merge operations on a key in the memtable.
This can be used to improve performance of the "get" operation. If many
successive merge operations are performed on a key, the performance of "get"
operations on the key deteriorates, as the value has to be computed for each
"get" operation by applying all the successive merge operations.
FB Task ID: #3428853
Test Plan:
make all check
db_bench --benchmarks=readrandommergerandom
counter_stress_test
Reviewers: haobo, vamsi, dhruba, sdong
Reviewed By: haobo
CC: zshao
Differential Revision: https://reviews.facebook.net/D14991
Summary:
For some tests I want to cache the database prior to running other tests on the same invocation
of db_bench. The readtocache test ignores --threads and --reads so those can be used by other tests
and it will still do a full read of --num rows with one thread. It might be invoked like:
db_bench --benchmarks=readtocache,readrandom --reads 100 --num 10000 --threads 8
Task ID: #
Blame Rev:
Test Plan:
run db_bench
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14739
Summary:
This adds 2 options for compression to db_bench:
* universal_compression_size_percent
* compression_level - to set zlib compression level
It also logs compression_size_percent at startup in LOG
Task ID: #
Blame Rev:
Test Plan:
make check, run db_bench
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14439
Summary:
Let's get rid of TransformRep and it's children. We have confirmed that HashSkipListRep works better with multifeed, so there is no benefit to keeping this around.
This diff is mostly just deleting references to obsoleted functions. I also have a diff for fbcode that we'll need to push when we switch to new release.
I had to expose HashSkipListRepFactory in the client header files because db_impl.cc needs access to GetTransform() function for SanitizeOptions.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14397
Summary:
Provide a framework to profile a query in detail to figure out latency bottleneck. Currently, in Get(), Put() and iterators, 2-3 simple timing is used. We can easily add more profile counters to the framework later.
Test Plan: Enable this profiling in seveal existing tests.
Reviewers: haobo, dhruba, kailiu, emayanke, vamsi, igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14001
Conflicts:
table/merger.cc
Summary:
Provide a framework to profile a query in detail to figure out latency bottleneck. Currently, in Get(), Put() and iterators, 2-3 simple timing is used. We can easily add more profile counters to the framework later.
Test Plan: Enable this profiling in seveal existing tests.
Reviewers: haobo, dhruba, kailiu, emayanke, vamsi, igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14001
Summary: This diff invoves some more complicated issues in the posix environment.
Test Plan: works under mac os. will need to verify dev box.
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14061
Summary:
Broke the compile when I removed purge_log_after_memtable_flush.
sorrybus
Test Plan: make db_bench works now
Reviewers: haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14037
Summary:
Archive cleaning will still happen every WAL_ttl seconds
but archived logs will be deleted only if archive size
is greater then a WAL_size_limit value.
Empty archived logs will be deleted evety WAL_ttl.
Test Plan:
1. Unit tests pass.
2. Benchmark.
Reviewers: emayanke, dhruba, haobo, sdong, kailiu, igor
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13869
Summary:
Rocksdb can now support a uncompressed block cache, or a compressed
block cache or both. Lookups first look for a block in the
uncompressed cache, if it is not found only then it is looked up
in the compressed cache. If it is found in the compressed cache,
then it is uncompressed and inserted into the uncompressed cache.
It is possible that the same block resides in the compressed cache
as well as the uncompressed cache at the same time. Both caches
have their own individual LRU policy.
Test Plan: Unit test case attached.
Reviewers: kailiu, sdong, haobo, leveldb
Reviewed By: haobo
CC: xjin, haobo
Differential Revision: https://reviews.facebook.net/D12675
Summary: Converted db_stress, db_repl_stress and db_bench to use gflags
Test Plan: I tested by printing out all the flags from old and new versions. Tried defaults, + various combinations with "interesting flags". Also, tested by running db_crashtest.py and db_crashtest2.py.
Reviewers: emayanke, dhruba, haobo, kailiu, sdong
Reviewed By: emayanke
CC: leveldb, xjin
Differential Revision: https://reviews.facebook.net/D13581
Summary:
With this patch, when LRUCache.Insert() is called and the cache is full, it will first try to free up entries whose reference counter is 1 (would become 0 after remo\
ving from the cache). We do it in two passes, in the first pass, we only try to release those unreferenced entries. If we cannot free enough space after traversing t\
he first remove_scan_cnt_ entries, we start from the beginning again and remove those entries being used.
Test Plan: add two unit tests to cover the codes
Reviewers: dhruba, haobo, emayanke
Reviewed By: emayanke
CC: leveldb, emayanke, xjin
Differential Revision: https://reviews.facebook.net/D13377
Summary:
Change namespace from leveldb to rocksdb. This allows a single
application to link in open-source leveldb code as well as
rocksdb code into the same process.
Test Plan: compile rocksdb
Reviewers: emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13287
Summary:
There is a use-case where we want to insert data into rocksdb as
fast as possible. Vector rep is used for this purpose.
The background flush thread needs to flush the vectorrep to
storage. It acquires the dblock then sorts the vector, releases
the dblock and then writes the sorted vector to storage. This is
suboptimal because the lock is held during the sort, which
prevents new writes for occuring.
This patch moves the sorting of the vector rep to outside the
db mutex. Performance is now as fastas the underlying storage
system. If you are doing buffered writes to rocksdb files, then
you can observe throughput upwards of 200 MB/sec writes.
This is an early draft and not yet ready to be reviewed.
Test Plan:
make check
Task ID: #
Blame Rev:
Reviewers: haobo
Reviewed By: haobo
CC: leveldb, haobo
Differential Revision: https://reviews.facebook.net/D12987
Summary:
Added a new field called max_size_amplification_ratio in the
CompactionOptionsUniversal structure. This determines the maximum
percentage overhead of space amplification.
The size amplification is defined to be the ratio between the size of
the oldest file to the sum of the sizes of all other files. If the
size amplification exceeds the specified value, then min_merge_width
and max_merge_width are ignored and a full compaction of all files is done.
A value of 10 means that the size a database that stores 100 bytes
of user data could occupy 110 bytes of physical storage.
Test Plan: Unit test DBTest.UniversalCompactionSpaceAmplification added.
Reviewers: haobo, emayanke, xjin
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12825
Summary: As title. The DB log file life cycle is tied up with the memtable it backs. Once the memtable is flushed to sst and committed, we should be able to delete the log file, without holding the mutex. This is part of the bigger change to avoid FindObsoleteFiles at runtime. It deals with log files. sst files will be dealt with later.
Test Plan: make check; db_bench
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11709
Summary: Fix code so that the filter_block layer only assumes keys are internal when prefix_extractor is set.
Test Plan: ./filter_block_test
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12501
Summary: Replace include/leveldb with include/rocksdb.
Test Plan:
make clean; make check
make clean; make release
Differential Revision: https://reviews.facebook.net/D12489
Summary:
This patch adds three new MemTableRep's: UnsortedRep, PrefixHashRep, and VectorRep.
UnsortedRep stores keys in an std::unordered_map of std::sets. When an iterator is requested, it dumps the keys into an std::set and iterates over that.
VectorRep stores keys in an std::vector. When an iterator is requested, it creates a copy of the vector and sorts it using std::sort. The iterator accesses that new vector.
PrefixHashRep stores keys in an unordered_map mapping prefixes to ordered sets.
I also added one API change. I added a function MemTableRep::MarkImmutable. This function is called when the rep is added to the immutable list. It doesn't do anything yet, but it seems like that could be useful. In particular, for the vectorrep, it means we could elide the extra copy and just sort in place. The only reason I haven't done that yet is because the use of the ArenaAllocator complicates things (I can elaborate on this if needed).
Test Plan:
make -j32 check
./db_stress --memtablerep=vector
./db_stress --memtablerep=unsorted
./db_stress --memtablerep=prefixhash --prefix_size=10
Reviewers: dhruba, haobo, emayanke
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12117
Summary: If use_prefix_filters is set and read_range>1, then the random seeks will set a the prefix filter to be the prefix of the key which was randomly selected as the target. Still need to add statistics (perhaps in a separate diff).
Test Plan: ./db_bench --benchmarks=fillseq,prefixscanrandom --num=10000000 --statistics=1 --use_prefix_blooms=1 --use_prefix_api=1 --bloom_bits=10
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb, haobo
Differential Revision: https://reviews.facebook.net/D12273
Test Plan:
- make all check;
- make release;
- make stringappend_test; ./stringappend_test
Reviewers: haobo, emayanke
Reviewed By: haobo
CC: leveldb, kailiu
Differential Revision: https://reviews.facebook.net/D12381
Summary:
In release, "found variable assigned but not used anywhere". Changed it to work with
assert. Someone accept this :).
Test Plan: make release -j 32
Reviewers: haobo, dhruba, emayanke
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12309
Summary:
Updated db_bench and utilities/merge_operators.h to allow for dynamic benchmarking
of merge operators in db_bench. Added a new test (--benchmarks=mergerandom), which performs
a bunch of random Merge() operations over random keys. Also added a "--merge_operator=" flag
so that the tester can easily benchmark different merge operators. Currently supports
the PutOperator and UInt64Add operator. Support for stringappend or list append may come later.
Test Plan:
1. make db_bench
2. Test the PutOperator (simulating Put) as follows:
./db_bench --benchmarks=fillrandom,readrandom,updaterandom,readrandom,mergerandom,readrandom --merge_operator=put
--threads=2
3. Test the UInt64AddOperator (simulating numeric addition) similarly:
./db_bench --value_size=8 --benchmarks=fillrandom,readrandom,updaterandom,readrandom,mergerandom,readrandom
--merge_operator=uint64add --threads=2
Reviewers: haobo, dhruba, zshao, MarkCallaghan
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11535
Summary:
Minor fix to current codes, including: coding style, output format,
comments. No major logic change. There are only 2 real changes, please see my inline comments.
Test Plan: make all check
Reviewers: haobo, dhruba, emayanke
Differential Revision: https://reviews.facebook.net/D12297
Summary:
This diff adds support for both soft and hard rate limiting. The following changes are included:
1) Options.rate_limit is renamed to Options.hard_rate_limit.
2) Options.rate_limit_delay_milliseconds is renamed to Options.rate_limit_delay_max_milliseconds.
3) Options.soft_rate_limit is added.
4) If the maximum compaction score is > hard_rate_limit and rate_limit_delay_max_milliseconds == 0, then writes are delayed by 1 ms at a time until the max compaction score falls below hard_rate_limit.
5) If the max compaction score is > soft_rate_limit but <= hard_rate_limit, then writes are delayed by 0-1 ms depending on how close we are to hard_rate_limit.
6) Users can disable 4 by setting hard_rate_limit = 0. They can add a limit to the maximum amount of time waited by setting rate_limit_delay_max_milliseconds > 0. Thus, the old behavior can be preserved by setting soft_rate_limit = 0, which is the default.
Test Plan:
make -j32 check
./db_stress
Reviewers: dhruba, haobo, MarkCallaghan
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12003
Summary:
Introduced KeyMayExist checking during writebatch-delete and removed from Outer Delete API because it uses writebatch-delete.
Added code to skip getting Table from disk if not already present in table_cache.
Some renaming of variables.
Introduced KeyMayExistImpl which allows checking since specified sequence number in GetImpl useful to check partially written writebatch.
Changed KeyMayExist to not be pure virtual and provided a default implementation.
Expanded unit-tests in db_test to check appropriately.
Ran db_stress for 1 hour with ./db_stress --max_key=100000 --ops_per_thread=10000000 --delpercent=50 --filter_deletes=1 --statistics=1.
Test Plan: db_stress;make check
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb, xjin
Differential Revision: https://reviews.facebook.net/D11745
Summary:
Wrote a new function in db_impl.c-CheckKeyMayExist that calls Get but with a new parameter turned on which makes Get return false only if bloom filters can guarantee that key is not in database. Delete calls this function and if the option- deletes_use_filter is turned on and CheckKeyMayExist returns false, the delete will be dropped saving:
1. Put of delete type
2. Space in the db,and
3. Compaction time
Test Plan:
make all check;
will run db_stress and db_bench and enhance unit-test once the basic design gets approved
Reviewers: dhruba, haobo, vamsi
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11607
Summary:
There is a new option called hybrid_mode which, when switched on,
causes HBase style compactions. Files from L0 are
compacted back into L0. This meat of this compaction algorithm
is in PickCompactionHybrid().
All files reside in L0. That means all files have overlapping
keys. Each file has a time-bound, i.e. each file contains a
range of keys that were inserted around the same time. The
start-seqno and the end-seqno refers to the timeframe when
these keys were inserted. Files that have contiguous seqno
are compacted together into a larger file. All files are
ordered from most recent to the oldest.
The current compaction algorithm starts to look for
candidate files starting from the most recent file. It continues to
add more files to the same compaction run as long as the
sum of the files chosen till now is smaller than the next
candidate file size. This logic needs to be debated
and validated.
The above logic should reduce write amplification to a
large extent... will publish numbers shortly.
Test Plan: dbstress runs for 6 hours with no data corruption (tested so far).
Differential Revision: https://reviews.facebook.net/D11289
Summary:
There is a new option called hybrid_mode which, when switched on,
causes HBase style compactions. Files from L0 are
compacted back into L0. This meat of this compaction algorithm
is in PickCompactionHybrid().
All files reside in L0. That means all files have overlapping
keys. Each file has a time-bound, i.e. each file contains a
range of keys that were inserted around the same time. The
start-seqno and the end-seqno refers to the timeframe when
these keys were inserted. Files that have contiguous seqno
are compacted together into a larger file. All files are
ordered from most recent to the oldest.
The current compaction algorithm starts to look for
candidate files starting from the most recent file. It continues to
add more files to the same compaction run as long as the
sum of the files chosen till now is smaller than the next
candidate file size. This logic needs to be debated
and validated.
The above logic should reduce write amplification to a
large extent... will publish numbers shortly.
Test Plan: dbstress runs for 6 hours with no data corruption (tested so far).
Differential Revision: https://reviews.facebook.net/D11289
Summary: This diff added an option to control the incremenal sync frequency. db_bench has a new flag bytes_per_sync for easy tuning exercise.
Test Plan: make check; db_bench
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11295
Summary:
simplify the printing code in db_bench
use TickersMap and HistogramsNameMap introduced in previous diffs.
Test Plan: ./db_bench --statistics=1 and see if all the statistics are printed
Reviewers: haobo, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11355
Summary:
Merge multiple multiple memtables in memory before writing it
out to a file in L0.
There is a new config parameter min_write_buffer_number_to_merge
that specifies the number of write buffers that should be merged
together to a single file in storage. The system will not flush
wrte buffers to storage unless at least these many buffers have
accumulated in memory.
The default value of this new parameter is 1, which means that
a write buffer will be immediately flushed to disk as soon it is
ready.
Test Plan: make check
Differential Revision: https://reviews.facebook.net/D11241
Igor Canadi