Commit Graph

21 Commits

Author SHA1 Message Date
Ari Ekmekji
40c64434d4 Parallelize L0-L1 Compaction: Restructure Compaction Job
Summary:
As of now compactions involving files from Level 0 and Level 1 are single
threaded because the files in L0, although sorted, are not range partitioned like
the other levels. This means that during L0-L1 compaction each file from L1
needs to be merged with potentially all the files from L0.

This attempt to parallelize the L0-L1 compaction assigns a thread and a
corresponding iterator to each L1 file that then considers only the key range
found in that L1 file and only the L0 files that have those keys (and only the
specific portion of those L0 files in which those keys are found). In this way
the overlap is minimized and potentially eliminated between different iterators
focusing on the same files.

The first step is to restructure the compaction logic to break L0-L1 compactions
into multiple, smaller, sequential compactions. Eventually each of these smaller
jobs will be run simultaneously. Areas to pay extra attention to are

  # Correct aggregation of compaction job statistics across multiple threads
  # Proper opening/closing of output files (make sure each thread's is unique)
  # Keys that span multiple L1 files
  # Skewed distributions of keys within L0 files

Test Plan: Make and run db_test (newer version has separate compaction tests) and compaction_job_stats_test

Reviewers: igor, noetzli, anthony, sdong, yhchiang

Reviewed By: yhchiang

Subscribers: MarkCallaghan, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D42699
2015-08-03 11:32:14 -07:00
Igor Canadi
35ca59364c Don't let flushes preempt compactions
Summary:
When we first started, max_background_flushes was 0 by default and compaction thread was executing flushes (since there was no flush thread). Then, we switched the default max_background_flushes to 1. However, we still support the case where there is no flush thread and flushes are done in compaction. This is making our code a bit more complicated. By not supporting this use-case we can make our code simpler.

We have a special case that when you set max_background_flushes to 0, we
schedule the flush to execute on the compaction thread.

Test Plan: make check (there might be some unit tests that depend on this behavior)

Reviewers: IslamAbdelRahman, yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D41931
2015-07-17 12:02:52 -07:00
Igor Canadi
a96fcd09b7 Deprecate CompactionFilterV2
Summary: It has been around for a while and it looks like it never found any uses in the wild. It's also complicating our compaction_job code quite a bit. We're deprecating it in 3.13, but will put it back in 3.14 if we actually find users that need this feature.

Test Plan: make check

Reviewers: noetzli, yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D42405
2015-07-17 18:59:11 +02:00
Andres Notzli
6b2d44b2ff Refactoring of writing key/value pairs
Summary:
Before, writing key/value pairs out to files was done inside
ProcessKeyValueCompaction(). To make ProcessKeyValueCompaction()
more understandable, this patch moves the writing part to a separate
function. This is intended to be a stepping stone for additional
changes.

Test Plan: make && make check

Reviewers: sdong, rven, yhchiang, igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D42243
2015-07-15 09:55:45 -07:00
Andres Notzli
ab137af4ba Partial cleanup of CompactionJob
Summary:
Logging, dealing with key prefix batches and updating stats
moved from CompactionJob::Run() into separate functions.

Test Plan: make all && make check

Reviewers: sdong, rven, yhchiang, igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D41919
2015-07-14 00:09:20 -07:00
Yueh-Hsuan Chiang
bb1c74ce18 Fixed a bug of CompactionStats in multi-level universal compaction case
Summary:
Universal compaction can involves in multiple levels.  However,
the current implementation of bytes_readn and bytes_readnp1
(and some other stats with postfix `n` and `np1`) assumes compaction
can only have two levels.

This patch fixes this bug and redefines bytes_readn and bytes_readnp1:
* bytes_readnp1: the number of bytes read in the compaction output level.
* bytes_readn: the total number of bytes read minus bytes_readnp1

Test Plan: Add a test in compaction_job_stats_test

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

Reviewed By: IslamAbdelRahman

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D40239
2015-06-17 23:40:34 -07:00
Yueh-Hsuan Chiang
fe5c6321cb Allow EventListener::OnCompactionCompleted to return CompactionJobStats.
Summary:
Allow EventListener::OnCompactionCompleted to return CompactionJobStats,
which contains useful information about a compaction.

Example CompactionJobStats returned by OnCompactionCompleted():
    smallest_output_key_prefix 05000000
    largest_output_key_prefix 06990000
    elapsed_time 42419
    num_input_records 300
    num_input_files 3
    num_input_files_at_output_level 2
    num_output_records 200
    num_output_files 1
    actual_bytes_input 167200
    actual_bytes_output 110688
    total_input_raw_key_bytes 5400
    total_input_raw_value_bytes 300000
    num_records_replaced 100
    is_manual_compaction 1

Test Plan: Developed a mega test in db_test which covers 20 variables in CompactionJobStats.

Reviewers: rven, igor, anthony, sdong

Reviewed By: sdong

Subscribers: tnovak, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D38463
2015-06-02 17:07:16 -07:00
Yueh-Hsuan Chiang
fc83821270 Add EventListener::OnTableFileCreated()
Summary:
Add EventListener::OnTableFileCreated(), which will be called
when a table file is created.  This patch is part of the
EventLogger and EventListener integration.

Test Plan: Augment existing test in db/listener_test.cc

Reviewers: anthony, kradhakrishnan, rven, igor, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D38865
2015-06-02 14:12:23 -07:00
Yueh-Hsuan Chiang
77a5a543a5 Allow GetThreadList() to report basic compaction operation properties.
Summary:
Now we're able to show more details about a compaction in
GetThreadList() :)

This patch allows GetThreadList() to report basic compaction
operation properties.  Basic compaction properties include:
    1. job id
    2. compaction input / output level
    3. compaction property flags (is_manual, is_deletion, .. etc)
    4. total input bytes
    5. the number of bytes has been read currently.
    6. the number of bytes has been written currently.

Flush operation properties will be done in a seperate diff.

Test Plan:
/db_bench --threads=30 --num=1000000 --benchmarks=fillrandom --thread_status_per_interval=1

Sample output of tracking same job:

          ThreadID ThreadType       cfName            Operation   ElapsedTime                                         Stage        State OperationProperties
   140664171987072    Low Pri      default           Compaction     31.357 ms     CompactionJob::FinishCompactionOutputFile              BaseInputLevel 1 | BytesRead 2264663 | BytesWritten 1934241 | IsDeletion 0 | IsManual 0 | IsTrivialMove 0 | JobID 277 | OutputLevel 2 | TotalInputBytes 3964158 |

          ThreadID ThreadType       cfName            Operation   ElapsedTime                                         Stage        State OperationProperties
   140664171987072    Low Pri      default           Compaction     59.440 ms     CompactionJob::FinishCompactionOutputFile              BaseInputLevel 1 | BytesRead 2264663 | BytesWritten 1934241 | IsDeletion 0 | IsManual 0 | IsTrivialMove 0 | JobID 277 | OutputLevel 2 | TotalInputBytes 3964158 |

          ThreadID ThreadType       cfName            Operation   ElapsedTime                                         Stage        State OperationProperties
   140664171987072    Low Pri      default           Compaction    226.375 ms                        CompactionJob::Install              BaseInputLevel 1 | BytesRead 3958013 | BytesWritten 3621940 | IsDeletion 0 | IsManual 0 | IsTrivialMove 0 | JobID 277 | OutputLevel 2 | TotalInputBytes 3964158 |

Reviewers: sdong, rven, igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D37653
2015-05-06 22:51:06 -07:00
Igor Canadi
65fe1cfbb3 Cleanup CompactionJob
Summary:
Couple changes:
1. instead of SnapshotList, just take a vector of snapshots
2. don't take a separate parameter is_snapshots_supported. If there are snapshots in the list, that means they are supported. I actually think we should get rid of this notion of snapshots not being supported.
3. don't pass in mutable_cf_options as a parameter. Lifetime of mutable_cf_options is a bit tricky to maintain, so it's better to not pass it in for the whole compaction job. We only really need it when we install the compaction results.

Test Plan: make check

Reviewers: sdong, rven, yhchiang

Reviewed By: yhchiang

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D36627
2015-05-05 19:01:12 -07:00
Igor Canadi
1bb4928da9 Include bunch of more events into EventLogger
Summary:
Added these events:
* Recovery start, finish and also when recovery creates a file
* Trivial move
* Compaction start, finish and when compaction creates a file
* Flush start, finish

Also includes small fix to EventLogger

Also added option ROCKSDB_PRINT_EVENTS_TO_STDOUT which is useful when we debug things. I've spent far too much time chasing LOG files.

Still didn't get sst table properties in JSON. They are written very deeply into the stack. I'll address in separate diff.

TODO:
* Write specification. Let's first use this for a while and figure out what's good data to put here, too. After that we'll write spec
* Write tools that parse and analyze LOGs. This can be in python or go. Good intern task.

Test Plan: Ran db_bench with ROCKSDB_PRINT_EVENTS_TO_STDOUT. Here's the output: https://phabricator.fb.com/P19811976

Reviewers: sdong, yhchiang, rven, MarkCallaghan, kradhakrishnan, anthony

Reviewed By: anthony

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D37521
2015-04-27 15:20:02 -07:00
Anurag Indu
3d1a924ff3 Adding stats for the merge and filter operation
Summary:
We have addded new stats and perf_context for measuring the merge and filter operation time consumption.
We have bounded all the merge operations within the GUARD statment and collected the total time for these operations in the DB.

Test Plan: WIP

Reviewers: rven, yhchiang, kradhakrishnan, igor, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D34377
2015-03-24 14:42:04 -07:00
Yueh-Hsuan Chiang
c594b0e89d Allow GetThreadList() to report operation stage.
Summary: Allow GetThreadList() to report operation stage.

Test Plan:
  ./thread_list_test
  ./db_bench --benchmarks=fillrandom --num=100000 --threads=40 \
    --max_background_compactions=10 --max_background_flushes=3 \
    --thread_status_per_interval=1000 --key_size=16 --value_size=1000 \
    --num_column_families=10

  export ROCKSDB_TESTS=ThreadStatus
  ./db_test

Sample output
          ThreadID ThreadType                    cfName    Operation        OP_StartTime    ElapsedTime                                         Stage        State
   140116265861184    Low Pri
   140116270055488    Low Pri
   140116274249792   High Pri column_family_name_000005        Flush 2015/03/10-14:58:11           0 us                    FlushJob::WriteLevel0Table
   140116400078912    Low Pri column_family_name_000004   Compaction 2015/03/10-14:58:11           0 us     CompactionJob::FinishCompactionOutputFile
   140116358135872    Low Pri column_family_name_000006   Compaction 2015/03/10-14:58:10           1 us     CompactionJob::FinishCompactionOutputFile
   140116341358656    Low Pri
   140116295221312   High Pri                   default        Flush 2015/03/10-14:58:11           0 us                    FlushJob::WriteLevel0Table
   140116324581440    Low Pri column_family_name_000009   Compaction 2015/03/10-14:58:11           0 us      CompactionJob::ProcessKeyValueCompaction
   140116278444096    Low Pri
   140116299415616    Low Pri column_family_name_000008   Compaction 2015/03/10-14:58:11           0 us     CompactionJob::FinishCompactionOutputFile
   140116291027008   High Pri column_family_name_000001        Flush 2015/03/10-14:58:11           0 us                    FlushJob::WriteLevel0Table
   140116286832704    Low Pri column_family_name_000002   Compaction 2015/03/10-14:58:11           0 us     CompactionJob::FinishCompactionOutputFile
   140116282638400    Low Pri

Reviewers: rven, igor, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D34683
2015-03-13 10:45:40 -07:00
Igor Canadi
e7ea51a8e7 Introduce job_id for flush and compaction
Summary:
It would be good to assing background job their IDs. Two benefits:
1) makes LOGs more readable
2) I might use it in my EventLogger, which will try to make our LOG easier to read/query/visualize

Test Plan: ran rocksdb, read the LOG

Reviewers: sdong, rven, yhchiang

Reviewed By: yhchiang

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D31617
2015-02-12 09:54:48 -08:00
Yueh-Hsuan Chiang
181191a1e4 Add a counter for collecting the wait time on db mutex.
Summary:
Add a counter for collecting the wait time on db mutex.
Also add MutexWrapper and CondVarWrapper for measuring wait time.

Test Plan:
./db_test
export ROCKSDB_TESTS=MutexWaitStats
./db_test

verify stats output using db_bench
make clean
make release
./db_bench --statistics=1 --benchmarks=fillseq,readwhilewriting --num=10000 --threads=10

Sample output:
    rocksdb.db.mutex.wait.micros COUNT : 7546866

Reviewers: MarkCallaghan, rven, sdong, igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D32787
2015-02-04 21:39:45 -08:00
sdong
d888c95748 Sync WAL Directory and DB Path if different from DB directory
Summary:
1. If WAL directory is different from db directory. Sync the directory after creating a log file under it.
2. After creating an SST file, sync its parent directory instead of DB directory.
3. change the check of kResetDeleteUnsyncedFiles in fault_injection_test. Since we changed the behavior to sync log files' parent directory after first WAL sync, instead of creating, kResetDeleteUnsyncedFiles will not guarantee to show post sync updates.

Test Plan: make all check

Reviewers: yhchiang, rven, igor

Reviewed By: igor

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D32067
2015-01-26 14:17:45 -08:00
Igor Canadi
4a3bd2bad2 Optimize usage of Status in CompactionJob
Summary: Based on @ljin feedback

Test Plan: compiles

Reviewers: ljin, yhchiang, sdong

Reviewed By: sdong

Subscribers: ljin, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D28515
2014-11-10 11:57:58 -08:00
Igor Canadi
e3d3567b5b Get rid of mutex in CompactionJob's state
Summary: Based on @sdong's feedback in the diff, we shouldn't keep db_mutex in CompactionJob's state. This diff removes db_mutex from CompactionJob state, by making next_file_number_ atomic. That way we only need to pass the lock to InstallCompactionResults() because of LogAndApply()

Test Plan: make check

Reviewers: ljin, yhchiang, rven, sdong

Reviewed By: sdong

Subscribers: sdong, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D28491
2014-11-07 15:44:12 -08:00
Igor Canadi
53af5d877d Redesign pending_outputs_
Summary:
Here's a prototype of redesigning pending_outputs_. This way, we don't have to expose pending_outputs_ to other classes (CompactionJob, FlushJob, MemtableList). DBImpl takes care of it.

Still have to write some comments, but should be good enough to start the discussion.

Test Plan: make check, will also run stress test

Reviewers: ljin, sdong, rven, yhchiang

Reviewed By: yhchiang

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D28353
2014-11-07 11:50:34 -08:00
sdong
2ea1219eb6 Fix RecordIn and RecordDrop stats
Summary:
1. fix possible overflow of the two stats by using uint64_t
2. use a similar source of data to calculate RecordDrop. Previous one is not correct.

Test Plan: See outputs of db_bench settings, and the results look reasonable

Reviewers: MarkCallaghan, ljin, igor

Reviewed By: igor

Subscribers: rven, leveldb, yhchiang, dhruba

Differential Revision: https://reviews.facebook.net/D28155
2014-11-05 11:03:34 -08:00
Igor Canadi
74eb4fbe93 CompactionJob
Summary:
Long awaited CompactionJob class! Move most compaction-related things from DBImpl to CompactionJob, making CompactionJob easier to test and understand.

Currently this is just replicating exactly the same functionality with as little as change as possible. As future work, we should:
1. Add CompactionJob tests (I think I'll do that tomorrow)
2. Reduce CompactionJob's state that it inherits from DBImpl
3. Figure out how to do yielding to flush better. Currently I implemented a callback as we agreed yesterday, but I don't think it's a good long term solution.

This reduces db_impl.cc from 5000+ LOC to 3400!

Test Plan: make check, will add CompactionJob-specific tests, probably also move some tests from db_test to compaction_job_test

Reviewers: rven, yhchiang, sdong, ljin

Reviewed By: ljin

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D27957
2014-10-31 16:31:25 -07:00