Summary:
When paranoid_checks is on, DBImpl::CheckConsistency() iterates over all sst files and calls Env::GetFileSize() for each of them. As far as I could understand, this is pretty arbitrary and doesn't affect correctness - if filesystem doesn't corrupt fsynced files, the file sizes will always match; if it does, it may as well corrupt contents as well as sizes, and rocksdb doesn't check contents on open.
If there are thousands of sst files, getting all their sizes takes a while. If, on top of that, Env is overridden to use some remote storage instead of local filesystem, it can be *really* slow and overload the remote storage service. This PR adds an option to not do GetFileSize(); instead it does GetChildren() for parent directory to check that all the expected sst files are at least present, but doesn't check their sizes.
We can't just disable paranoid_checks instead because paranoid_checks do a few other important things: make the DB read-only on write errors, print error messages on read errors, etc.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6353
Test Plan: ran the added sanity check unit test. Will try it out in a LogDevice test cluster where the GetFileSize() calls are causing a lot of trouble.
Differential Revision: D19656425
Pulled By: al13n321
fbshipit-source-id: c2c421b367633033760d1f56747bad206d1fbf82
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
Summary:
The max batch size that we can write to the WAL is controlled by a static manner. So if the leader write is less than 128 KB we will have the batch size as leader write size + 128 KB else the limit will be 1 MB. Both of them are statically defined.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5759
Differential Revision: D17329298
fbshipit-source-id: a3d910629d8d8ca84ea39ad89c2b2d284571ded5
Summary:
Each DB has a globally unique ID. A DB can be physically copied around, or backed-up and restored, and the users should be identify the same DB. This unique ID right now is stored as plain text in file IDENTITY under the DB directory. This approach introduces at least two problems: (1) the file is not checksumed; (2) the source of truth of a DB is the manifest file, which can be copied separately from IDENTITY file, causing the DB ID to be wrong.
The goal of this PR is solve this problem by moving the DB ID to manifest. To begin with we will write to both identity file and manifest. Write to Manifest is controlled via the flag write_dbid_to_manifest in Options and default is false.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5725
Test Plan: Added unit tests.
Differential Revision: D16963840
Pulled By: vjnadimpalli
fbshipit-source-id: 8a86a4c8c82c716003c40fd6b9d2d758030d92e9
Summary:
Added log_readahead_size option to control prefetching for Log::Reader.
This is mostly useful for reading a remotely located log, as it can save the number of round-trips when reading it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5592
Differential Revision: D16362989
Pulled By: elipoz
fbshipit-source-id: c5d4d5245a44008cd59879640efff70c091ad3e8
Summary:
This PR continues the work in https://github.com/facebook/rocksdb/pull/4748 and https://github.com/facebook/rocksdb/pull/4535 by adding a new DBOption `persist_stats_to_disk` which instructs RocksDB to persist stats history to RocksDB itself. When statistics is enabled, and both options `stats_persist_period_sec` and `persist_stats_to_disk` are set, RocksDB will periodically write stats to a built-in column family in the following form: key -> (timestamp in microseconds)#(stats name), value -> stats value. The existing API `GetStatsHistory` will detect the current value of `persist_stats_to_disk` and either read from in-memory data structure or from the hidden column family on disk.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5046
Differential Revision: D15863138
Pulled By: miasantreble
fbshipit-source-id: bb82abdb3f2ca581aa42531734ac799f113e931b
Summary:
Performing unordered writes in rocksdb when unordered_write option is set to true. When enabled the writes to memtable are done without joining any write thread. This offers much higher write throughput since the upcoming writes would not have to wait for the slowest memtable write to finish. The tradeoff is that the writes visible to a snapshot might change over time. If the application cannot tolerate that, it should implement its own mechanisms to work around that. Using TransactionDB with WRITE_PREPARED write policy is one way to achieve that. Doing so increases the max throughput by 2.2x without however compromising the snapshot guarantees.
The patch is prepared based on an original by siying
Existing unit tests are extended to include unordered_write option.
Benchmark Results:
```
TEST_TMPDIR=/dev/shm/ ./db_bench_unordered --benchmarks=fillrandom --threads=32 --num=10000000 -max_write_buffer_number=16 --max_background_jobs=64 --batch_size=8 --writes=3000000 -level0_file_num_compaction_trigger=99999 --level0_slowdown_writes_trigger=99999 --level0_stop_writes_trigger=99999 -enable_pipelined_write=false -disable_auto_compactions --unordered_write=1
```
With WAL
- Vanilla RocksDB: 78.6 MB/s
- WRITER_PREPARED with unordered_write: 177.8 MB/s (2.2x)
- unordered_write: 368.9 MB/s (4.7x with relaxed snapshot guarantees)
Without WAL
- Vanilla RocksDB: 111.3 MB/s
- WRITER_PREPARED with unordered_write: 259.3 MB/s MB/s (2.3x)
- unordered_write: 645.6 MB/s (5.8x with relaxed snapshot guarantees)
- WRITER_PREPARED with unordered_write disable concurrency control: 185.3 MB/s MB/s (2.35x)
Limitations:
- The feature is not yet extended to `max_successive_merges` > 0. The feature is also incompatible with `enable_pipelined_write` = true as well as with `allow_concurrent_memtable_write` = false.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5218
Differential Revision: D15219029
Pulled By: maysamyabandeh
fbshipit-source-id: 38f2abc4af8780148c6128acdba2b3227bc81759
Summary:
The existing implementation does not guarantee bytes reach disk every `bytes_per_sync` when writing SST files, or every `wal_bytes_per_sync` when writing WALs. This can cause confusing behavior for users who enable this feature to avoid large syncs during flush and compaction, but then end up hitting them anyways.
My understanding of the existing behavior is we used `sync_file_range` with `SYNC_FILE_RANGE_WRITE` to submit ranges for async writeback, such that we could continue processing the next range of bytes while that I/O is happening. I believe we can preserve that benefit while also limiting how far the processing can get ahead of the I/O, which prevents huge syncs from happening when the file finishes.
Consider this `sync_file_range` usage: `sync_file_range(fd_, 0, static_cast<off_t>(offset + nbytes), SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE)`. Expanding the range to start at 0 and adding the `SYNC_FILE_RANGE_WAIT_BEFORE` flag causes any pending writeback (like from a previous call to `sync_file_range`) to finish before it proceeds to submit the latest `nbytes` for writeback. The latest `nbytes` are still written back asynchronously, unless processing exceeds I/O speed, in which case the following `sync_file_range` will need to wait on it.
There is a second change in this PR to use `fdatasync` when `sync_file_range` is unavailable (determined statically) or has some known problem with the underlying filesystem (determined dynamically).
The above two changes only apply when the user enables a new option, `strict_bytes_per_sync`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5183
Differential Revision: D14953553
Pulled By: siying
fbshipit-source-id: 445c3862e019fb7b470f9c7f314fc231b62706e9
Summary:
Just like ReadOptions::background_purge_on_iterator_cleanup but for ColumnFamilyHandle instead of Iterator.
In our use case we sometimes call ColumnFamilyHandle's destructor from low-latency threads, and sometimes it blocks the thread for a few seconds deleting the files. To avoid that, we can either offload ColumnFamilyHandle's destruction to a background thread on our side, or add this option on rocksdb side. This PR does the latter, to be consistent with how we solve exactly the same problem for iterators using background_purge_on_iterator_cleanup option.
(EDIT: It's avoid_unnecessary_blocking_io now, and affects both CF drops and iterator destructors.)
I'm not quite comfortable with having two separate options (background_purge_on_iterator_cleanup and background_purge_on_cf_cleanup) for such a rarely used thing. Maybe we should merge them? Rename background_purge_on_cf_cleanup to something like delete_files_on_background_threads_only or avoid_blocking_io_in_unexpected_places, and make iterators use it instead of the one in ReadOptions? I can do that here if you guys think it's better.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5043
Differential Revision: D14339233
Pulled By: al13n321
fbshipit-source-id: ccf7efa11c85c9a5b91d969bb55627d0fb01e7b8
Summary:
This PR adds public `GetStatsHistory` API to retrieve stats history in the form of an std map. The key of the map is the timestamp in microseconds when the stats snapshot is taken, the value is another std map from stats name to stats value (stored in std string). Two DBOptions are introduced: `stats_persist_period_sec` (default 10 minutes) controls the intervals between two snapshots are taken; `max_stats_history_count` (default 10) controls the max number of history snapshots to keep in memory. RocksDB will stop collecting stats snapshots if `stats_persist_period_sec` is set to 0.
(This PR is the in-memory part of https://github.com/facebook/rocksdb/pull/4535)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4748
Differential Revision: D13961471
Pulled By: miasantreble
fbshipit-source-id: ac836d401ecb84ea92216bf9966f969dedf4ad04
Summary:
The motivation for this PR is to add to RocksDB support for differential (incremental) snapshots, as snapshot of the DB changes between two points in time (one can think of it as diff between to sequence numbers, or the diff D which can be thought of as an SST file or just set of KVs that can be applied to sequence number S1 to get the database to the state at sequence number S2).
This feature would be useful for various distributed storages layers built on top of RocksDB, as it should help reduce resources (time and network bandwidth) needed to recover and rebuilt DB instances as replicas in the context of distributed storages.
From the API standpoint that would like client app requesting iterator between (start seqnum) and current DB state, and reading the "diff".
This is a very draft PR for initial review in the discussion on the approach, i'm going to rework some parts and keep updating the PR.
For now, what's done here according to initial discussions:
Preserving deletes:
- We want to be able to optionally preserve recent deletes for some defined period of time, so that if a delete came in recently and might need to be included in the next incremental snapshot it would't get dropped by a compaction. This is done by adding new param to Options (preserve deletes flag) and new variable to DB Impl where we keep track of the sequence number after which we don't want to drop tombstones, even if they are otherwise eligible for deletion.
- I also added a new API call for clients to be able to advance this cutoff seqnum after which we drop deletes; i assume it's more flexible to let clients control this, since otherwise we'd need to keep some kind of timestamp < -- > seqnum mapping inside the DB, which sounds messy and painful to support. Clients could make use of it by periodically calling GetLatestSequenceNumber(), noting the timestamp, doing some calculation and figuring out by how much we need to advance the cutoff seqnum.
- Compaction codepath in compaction_iterator.cc has been modified to avoid dropping tombstones with seqnum > cutoff seqnum.
Iterator changes:
- couple params added to ReadOptions, to optionally allow client to request internal keys instead of user keys (so that client can get the latest value of a key, be it delete marker or a put), as well as min timestamp and min seqnum.
TableCache changes:
- I modified table_cache code to be able to quickly exclude SST files from iterators heep if creation_time on the file is less then iter_start_ts as passed in ReadOptions. That would help a lot in some DB settings (like reading very recent data only or using FIFO compactions), but not so much for universal compaction with more or less long iterator time span.
What's left:
- Still looking at how to best plug that inside DBIter codepath. So far it seems that FindNextUserKeyInternal only parses values as UserKeys, and iter->key() call generally returns user key. Can we add new API to DBIter as internal_key(), and modify this internal method to optionally set saved_key_ to point to the full internal key? I don't need to store actual seqnum there, but I do need to store type.
Closes https://github.com/facebook/rocksdb/pull/2999
Differential Revision: D6175602
Pulled By: mikhail-antonov
fbshipit-source-id: c779a6696ee2d574d86c69cec866a3ae095aa900
Summary:
SUMMARY
Moves the bytes_per_sync and wal_bytes_per_sync options from immutableoptions to mutable options. Also if wal_bytes_per_sync is changed, the wal file and memtables are flushed.
TEST PLAN
ran make check
all passed
Two new tests SetBytesPerSync, SetWalBytesPerSync check that after issuing setoptions with a new value for the var, the db options have the new value.
Closes https://github.com/facebook/rocksdb/pull/2893
Reviewed By: yiwu-arbug
Differential Revision: D5845814
Pulled By: TheRushingWookie
fbshipit-source-id: 93b52d779ce623691b546679dcd984a06d2ad1bd
Summary:
By default the seq number in DB is increased once per written key. WritePrepared txns requires the seq to be increased once per the entire batch so that the seq would be used as the prepare timestamp by which the transaction is identified. Also we need to increase seq for the commit marker since it would give a unique id to the commit timestamp of transactions.
Two unit tests are added to verify our understanding of how the seq should be increased. The recovery path requires much more work and is left to another patch.
Closes https://github.com/facebook/rocksdb/pull/2885
Differential Revision: D5837843
Pulled By: maysamyabandeh
fbshipit-source-id: a08960b93d727e1cf438c254d0c2636fb133cc1c
Summary:
Throughput: 46k tps in our sysbench settings (filling the details later)
The idea is to have the simplest change that gives us a reasonable boost
in 2PC throughput.
Major design changes:
1. The WAL file internal buffer is not flushed after each write. Instead
it is flushed before critical operations (WAL copy via fs) or when
FlushWAL is called by MySQL. Flushing the WAL buffer is also protected
via mutex_.
2. Use two sequence numbers: last seq, and last seq for write. Last seq
is the last visible sequence number for reads. Last seq for write is the
next sequence number that should be used to write to WAL/memtable. This
allows to have a memtable write be in parallel to WAL writes.
3. BatchGroup is not used for writes. This means that we can have
parallel writers which changes a major assumption in the code base. To
accommodate for that i) allow only 1 WriteImpl that intends to write to
memtable via mem_mutex_--which is fine since in 2PC almost all of the memtable writes
come via group commit phase which is serial anyway, ii) make all the
parts in the code base that assumed to be the only writer (via
EnterUnbatched) to also acquire mem_mutex_, iii) stat updates are
protected via a stat_mutex_.
Note: the first commit has the approach figured out but is not clean.
Submitting the PR anyway to get the early feedback on the approach. If
we are ok with the approach I will go ahead with this updates:
0) Rebase with Yi's pipelining changes
1) Currently batching is disabled by default to make sure that it will be
consistent with all unit tests. Will make this optional via a config.
2) A couple of unit tests are disabled. They need to be updated with the
serial commit of 2PC taken into account.
3) Replacing BatchGroup with mem_mutex_ got a bit ugly as it requires
releasing mutex_ beforehand (the same way EnterUnbatched does). This
needs to be cleaned up.
Closes https://github.com/facebook/rocksdb/pull/2345
Differential Revision: D5210732
Pulled By: maysamyabandeh
fbshipit-source-id: 78653bd95a35cd1e831e555e0e57bdfd695355a4
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:
PipelineWriteImpl is an alternative approach to WriteImpl. In WriteImpl, only one thread is allow to write at the same time. This thread will do both WAL and memtable writes for all write threads in the write group. Pending writers wait in queue until the current writer finishes. In the pipeline write approach, two queue is maintained: one WAL writer queue and one memtable writer queue. All writers (regardless of whether they need to write WAL) will still need to first join the WAL writer queue, and after the house keeping work and WAL writing, they will need to join memtable writer queue if needed. The benefit of this approach is that
1. Writers without memtable writes (e.g. the prepare phase of two phase commit) can exit write thread once WAL write is finish. They don't need to wait for memtable writes in case of group commit.
2. Pending writers only need to wait for previous WAL writer finish to be able to join the write thread, instead of wait also for previous memtable writes.
Merging #2056 and #2058 into this PR.
Closes https://github.com/facebook/rocksdb/pull/2286
Differential Revision: D5054606
Pulled By: yiwu-arbug
fbshipit-source-id: ee5b11efd19d3e39d6b7210937b11cefdd4d1c8d
Summary:
First cut for early review; there are few conceptual points to answer and some code structure issues.
For conceptual points -
- restriction-wise, we're going to disallow ingest_behind if (use_seqno_zero_out=true || disable_auto_compaction=false), the user is responsible to properly open and close DB with required params
- we wanted to ingest into reserved bottom most level. Should we fail fast if bottom level isn't empty, or should we attempt to ingest if file fits there key-ranges-wise?
- Modifying AssignLevelForIngestedFile seems the place we we'd handle that.
On code structure - going to refactor GenerateAndAddExternalFile call in the test class to allow passing instance of IngestionOptions, that's just going to incur lots of changes at callsites.
Closes https://github.com/facebook/rocksdb/pull/2144
Differential Revision: D4873732
Pulled By: lightmark
fbshipit-source-id: 81cb698106b68ef8797f564453651d50900e153a
Summary:
Makes max_open_files db option dynamically set-able by SetDBOptions. During the call of SetDBOptions we call SetCapacity on the table cache, which is a LRUCache.
Closes https://github.com/facebook/rocksdb/pull/2185
Differential Revision: D4979189
Pulled By: yiwu-arbug
fbshipit-source-id: ca7e8dc5e3619c79434f579be4847c0f7e56afda
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:
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:
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
Summary:
Made delete_obsolete_files_period_micros option dynamic. It can be updating using DB::SetDBOptions().
Closes https://github.com/facebook/rocksdb/pull/1595
Differential Revision: D4246569
Pulled By: tonek
fbshipit-source-id: d23f560
Summary: Use ImmutableDBOptions/MutableDBOptions internally and DBOptions only for user-facing APIs. MutableDBOptions is barely a placeholder for now. I'll start to move options to MutableDBOptions in following diffs.
Test Plan:
make all check
Reviewers: yhchiang, IslamAbdelRahman, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D64065
Mike Kolupaev