Summary:
Improve write buffer manager in several ways:
1. Size is tracked when arena block is allocated, rather than every allocation, so that it can better track actual memory usage and the tracking overhead is slightly lower.
2. We start to trigger memtable flush when 7/8 of the memory cap hits, instead of 100%, and make 100% much harder to hit.
3. Allow a cache object to be passed into buffer manager and the size allocated by memtable can be costed there. This can help users have one single memory cap across block cache and memtable.
Closes https://github.com/facebook/rocksdb/pull/2350
Differential Revision: D5110648
Pulled By: siying
fbshipit-source-id: b4238113094bf22574001e446b5d88523ba00017
Summary:
fix regression test by not reporting stats when building db
Closes https://github.com/facebook/rocksdb/pull/2390
Differential Revision: D5159909
Pulled By: lightmark
fbshipit-source-id: c3f4b9deb9c6799ff84207fd341c529144f8158d
Summary:
- `max_background_flushes` and `max_background_compactions` are still supported for backwards compatibility
- `base_background_compactions` is completely deprecated. Now we just throttle to one background compaction when there's no pressure.
- `max_background_jobs` is added to automatically partition the concurrent background jobs into flushes vs compactions. Currently it's very simple as we just allocate one-fourth of the jobs to flushes, and the remaining can be used for compactions.
- The test cases that set `base_background_compactions > 1` needed to be updated. I just grab the pressure token such that the desired number of compactions can be scheduled.
Closes https://github.com/facebook/rocksdb/pull/2205
Differential Revision: D4937461
Pulled By: ajkr
fbshipit-source-id: df52cbbd497e13bbc9a60560a5ac2a2526b3f1f9
Summary:
Release builds are failing on Linux with the error:
```
tools/db_stress.cc: In function ‘int main(int, char**)’:
tools/db_stress.cc:2365:12: error: ‘rocksdb::SyncPoint’ has not been declared
rocksdb::SyncPoint::GetInstance()->SetCallBack(
^
tools/db_stress.cc:2370:12: error: ‘rocksdb::SyncPoint’ has not been declared
rocksdb::SyncPoint::GetInstance()->SetCallBack(
^
tools/db_stress.cc:2375:12: error: ‘rocksdb::SyncPoint’ has not been declared
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
^
make[1]: *** [tools/db_stress.o] Error 1
make[1]: Leaving directory `/data/sandcastle/boxes/trunk-git-rocksdb-public'
make: *** [release] Error 2
```
Closes https://github.com/facebook/rocksdb/pull/2355
Differential Revision: D5113552
Pulled By: sagar0
fbshipit-source-id: 351df707277787da5633ba4a40e52edc7c895dc4
Summary:
Disable direct reads for log and manifest. Direct reads should not affect sequential_file
Also add kDirectIO for option_config_ in db_test_util
Closes https://github.com/facebook/rocksdb/pull/2337
Differential Revision: D5100261
Pulled By: lightmark
fbshipit-source-id: 0ebfd13b93fa1b8f9acae514ac44f8125a05868b
Summary:
- Introduced an include/ file dedicated to db-related debug functions to avoid making db.h more complex
- Added debugging function, `GetAllKeyVersions()`, to return a listing of internal data for a range of user keys. The new `struct KeyVersion` exposes data similar to internal key without exposing any internal type.
- Migrated the "ldb idump" subcommand to use this function
- The API takes an inclusive-exclusive range to match behavior of "ldb idump". This will be quite annoying for users who want to query a single user key's versions :(.
Closes https://github.com/facebook/rocksdb/pull/2232
Differential Revision: D4976007
Pulled By: ajkr
fbshipit-source-id: cab375da53a7595d6575af2b7e3b776aa3ad793e
Summary:
Allow an option for users to do some compaction in FIFO compaction, to pay some write amplification for fewer number of files.
Closes https://github.com/facebook/rocksdb/pull/2163
Differential Revision: D4895953
Pulled By: siying
fbshipit-source-id: a1ab608dd0627211f3e1f588a2e97159646e1231
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
Summary:
In a previous commit, I changed the way to checkout release branches from "git checkout <branch_name>" to "git checkout origin/<branch_name>". However, this doesn't seem to work in our CI environment. Revert it.
Closes https://github.com/facebook/rocksdb/pull/2189
Differential Revision: D4922294
Pulled By: siying
fbshipit-source-id: 482c17f9b05e6ccb190876b050682fe5a458103d
Summary:
tools/check_format_compatible.sh will check a newer version of RocksDB can open option files generated by older version releases. In order to achieve that, a new parameter "--try_load_options" is added to ldb. With this parameter set, if option file exists, we load the option file and use it to open the DB. With this opiton set, we can validate option loading logic.
Closes https://github.com/facebook/rocksdb/pull/2178
Differential Revision: D4914989
Pulled By: siying
fbshipit-source-id: db114f7724fcb41e5e9483116d84d7c4b8389ca4
Summary:
Tested by running it on a remote machine.
I could not run it on the particular remote machine which has a different location for time command since it is busy and the script does not allow concurrent runs. So I tested it by hacking the script and replacing the command with "\$(hostname)" and confirmed that the scripts prints out the host name of the remote machine.
Closes https://github.com/facebook/rocksdb/pull/2181
Differential Revision: D4921654
Pulled By: maysamyabandeh
fbshipit-source-id: 8abb5ea9f7234f3c50a749576ccbb47ff605beb9
Summary:
Need to add more recent versions to tools/check_format_compatible.sh to meka sure backward and forward compatibility.
Closes https://github.com/facebook/rocksdb/pull/2175
Differential Revision: D4911585
Pulled By: siying
fbshipit-source-id: 943e6488757efb11bb6720d811c7ba949915c9de
Summary:
Add a function to allow users to reset internal stats without restarting the DB.
Closes https://github.com/facebook/rocksdb/pull/2167
Differential Revision: D4907939
Pulled By: siying
fbshipit-source-id: ab2dd85b88aabe9380da7485320a1d460d3e1f68
Summary:
Replace Options::use_direct_writes with Options::use_direct_io_for_flush_and_compaction
Now if Options::use_direct_io_for_flush_and_compaction = true, we will enable direct io for both reads and writes for flush and compaction job. Whereas Options::use_direct_reads controls user reads like iterator and Get().
Closes https://github.com/facebook/rocksdb/pull/2117
Differential Revision: D4860912
Pulled By: lightmark
fbshipit-source-id: d93575a8a5e780cf7e40797287edc425ee648c19
Summary:
filter the warning out and only print it once.
Closes https://github.com/facebook/rocksdb/pull/2137
Differential Revision: D4870925
Pulled By: lightmark
fbshipit-source-id: 91b363ce7f70bce88b0780337f408fc4649139b8
Summary:
Run the time command before regression tests, parse the output, and add the numbers to the report.
Closes https://github.com/facebook/rocksdb/pull/2101
Differential Revision: D4862781
Pulled By: maysamyabandeh
fbshipit-source-id: 4a81caa5d14187d67093aad154c8f0ad56aba901
Summary:
Check the result of the benchmark againt a specified truth_db, which is
expected to be produced using the same benchmark but perhaps on a
different commit or with different configs.
The verification is simple and assumes that key/values are generated
deterministically. This assumption would break if db_bench using rand
variable differently from the benchmark that produced truth_db.
Currently it is checked to work on fillrandom and readwhilewriting.
A param finish_after_writes is added to ensure that the background
writing thread will write the same number of entries between two
benchmarks.
Example:
$ TEST_TMPDIR=/dev/shm/truth_db ./db_bench
--benchmarks="fillrandom,readwhilewriting" --num=200000
--finish_after_writes=true
$ TEST_TMPDIR=/dev/shm/tmpdb ./db_bench
--benchmarks="fillrandom,readwhilewriting,verify" --truth_db
/dev/shm/truth_db/dbbench --num=200000 --finish_after_writes=true
Verifying db <= truth_db...
Verifying db >= truth_db...
...Verified
Closes https://github.com/facebook/rocksdb/pull/2098
Differential Revision: D4839233
Pulled By: maysamyabandeh
fbshipit-source-id: 2f4ed31
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
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
Summary:
Rebuilding regression db at the first run every monday.
Closes https://github.com/facebook/rocksdb/pull/2093
Differential Revision: D4836961
Pulled By: lightmark
fbshipit-source-id: 22d6c25
Summary:
It is confusing to have auto_roll_logger to stay under db/, which has nothing to do with database. Move filename together as it is a dependency.
Closes https://github.com/facebook/rocksdb/pull/2080
Differential Revision: D4821141
Pulled By: siying
fbshipit-source-id: ca7d768
Summary:
Fix the bug that previous test existence locally
Closes https://github.com/facebook/rocksdb/pull/2074
Differential Revision: D4813631
Pulled By: lightmark
fbshipit-source-id: eceb0d9
Summary:
add ldb to regression test in order to enable reuse db by creating checkpoint
Closes https://github.com/facebook/rocksdb/pull/2030
Differential Revision: D4800549
Pulled By: lightmark
fbshipit-source-id: d3a7325
Summary:
Added fifo benchmark to db_bench.
One thing i am not sure is that i am using CompactRange() instead of CompactFiles(). (may cause performance skew because CompactionRange() is not happening in current thread?) For CompactFiles(), for some reason FIFO compaction doesn't work as expected. More insight is welcomed. I guess FIFO compaction doesn't work with file names? igorcanadi
test cmd:
./db_bench --compaction_style=2 --benchmarks=fillseqdeterministic --disable_auto_compactions --num_levels=1 --fifo_compaction_max_table_files_size_mb=10
---------------------- DB 0 LSM ---------------------
Level[0]: /000014.sst(size: 4211014 bytes)
fillseqdeterministic : 4.731 micros/op 211381 ops/sec; 23.4 MB/s
Closes https://github.com/facebook/rocksdb/pull/1734
Differential Revision: D4774964
Pulled By: siying
fbshipit-source-id: 9d08df6
Summary:
PlainTable now supports non-mmap mode. We don't need to check it anymore.
Closes https://github.com/facebook/rocksdb/pull/1882
Differential Revision: D4751643
Pulled By: siying
fbshipit-source-id: ab14540
Summary:
Removed max_grandparent_overlap_factor from benchmark.sh since it is not a valid option anymore.
Closes https://github.com/facebook/rocksdb/pull/2015
Differential Revision: D4748229
Pulled By: lgalanis
fbshipit-source-id: c3869ea
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
Summary:
Add the flag --prefix to the sst_dump tool
This flag is similar to, and exclusive from, the --from flag.
--prefix=0x00FF will return all rows prefixed with 0x00FF.
The --to flag may also be specified and will work as expected.
These changes were used to help in debugging the power cycle corruption issue and theses changes were tested by scanning through a udb.
Closes https://github.com/facebook/rocksdb/pull/1984
Differential Revision: D4691814
Pulled By: reidHoruff
fbshipit-source-id: 027f261
Summary:
This reverts commit d43adf21bb.
The patch has caused problems in regression tests. Will revert it for now until we figure how to debug the problems regression tests.
Closes https://github.com/facebook/rocksdb/pull/1975
Differential Revision: D4682880
Pulled By: maysamyabandeh
fbshipit-source-id: 84df83a
Summary:
It augments the regression benchmarks with a time command, parses the output, and print them to the SUMMARY.csv file.
I tested a variation of the script locally. Any idea how to do run a test that also involves writing to scuba tables?
Closes https://github.com/facebook/rocksdb/pull/1967
Differential Revision: D4679470
Pulled By: maysamyabandeh
fbshipit-source-id: 44dac30
Siying Dong