You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
rocksdb/options/db_options.cc

302 lines
14 KiB

// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under both the GPLv2 (found in the
// COPYING file in the root directory) and Apache 2.0 License
// (found in the LICENSE.Apache file in the root directory).
#include "options/db_options.h"
#ifndef __STDC_FORMAT_MACROS
#define __STDC_FORMAT_MACROS
#endif
#include <inttypes.h>
#include "port/port.h"
#include "rocksdb/cache.h"
#include "rocksdb/env.h"
#include "rocksdb/sst_file_manager.h"
#include "rocksdb/wal_filter.h"
#include "util/logging.h"
namespace rocksdb {
ImmutableDBOptions::ImmutableDBOptions() : ImmutableDBOptions(Options()) {}
ImmutableDBOptions::ImmutableDBOptions(const DBOptions& options)
: create_if_missing(options.create_if_missing),
create_missing_column_families(options.create_missing_column_families),
error_if_exists(options.error_if_exists),
paranoid_checks(options.paranoid_checks),
env(options.env),
rate_limiter(options.rate_limiter),
sst_file_manager(options.sst_file_manager),
info_log(options.info_log),
info_log_level(options.info_log_level),
max_file_opening_threads(options.max_file_opening_threads),
statistics(options.statistics),
use_fsync(options.use_fsync),
db_paths(options.db_paths),
db_log_dir(options.db_log_dir),
wal_dir(options.wal_dir),
max_subcompactions(options.max_subcompactions),
max_background_flushes(options.max_background_flushes),
max_log_file_size(options.max_log_file_size),
log_file_time_to_roll(options.log_file_time_to_roll),
keep_log_file_num(options.keep_log_file_num),
recycle_log_file_num(options.recycle_log_file_num),
max_manifest_file_size(options.max_manifest_file_size),
table_cache_numshardbits(options.table_cache_numshardbits),
wal_ttl_seconds(options.WAL_ttl_seconds),
wal_size_limit_mb(options.WAL_size_limit_MB),
manifest_preallocation_size(options.manifest_preallocation_size),
allow_mmap_reads(options.allow_mmap_reads),
allow_mmap_writes(options.allow_mmap_writes),
use_direct_reads(options.use_direct_reads),
use_direct_io_for_flush_and_compaction(
options.use_direct_io_for_flush_and_compaction),
allow_fallocate(options.allow_fallocate),
is_fd_close_on_exec(options.is_fd_close_on_exec),
advise_random_on_open(options.advise_random_on_open),
db_write_buffer_size(options.db_write_buffer_size),
write_buffer_manager(options.write_buffer_manager),
access_hint_on_compaction_start(options.access_hint_on_compaction_start),
new_table_reader_for_compaction_inputs(
options.new_table_reader_for_compaction_inputs),
random_access_max_buffer_size(options.random_access_max_buffer_size),
use_adaptive_mutex(options.use_adaptive_mutex),
listeners(options.listeners),
enable_thread_tracking(options.enable_thread_tracking),
enable_pipelined_write(options.enable_pipelined_write),
allow_concurrent_memtable_write(options.allow_concurrent_memtable_write),
enable_write_thread_adaptive_yield(
options.enable_write_thread_adaptive_yield),
write_thread_max_yield_usec(options.write_thread_max_yield_usec),
write_thread_slow_yield_usec(options.write_thread_slow_yield_usec),
skip_stats_update_on_db_open(options.skip_stats_update_on_db_open),
wal_recovery_mode(options.wal_recovery_mode),
allow_2pc(options.allow_2pc),
row_cache(options.row_cache),
#ifndef ROCKSDB_LITE
wal_filter(options.wal_filter),
#endif // ROCKSDB_LITE
fail_if_options_file_error(options.fail_if_options_file_error),
dump_malloc_stats(options.dump_malloc_stats),
avoid_flush_during_recovery(options.avoid_flush_during_recovery),
Optimize for serial commits in 2PC 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
7 years ago
allow_ingest_behind(options.allow_ingest_behind),
Added support for differential snapshots 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
7 years ago
preserve_deletes(options.preserve_deletes),
two_write_queues(options.two_write_queues),
manual_wal_flush(options.manual_wal_flush),
atomic_flush(options.atomic_flush),
avoid_unnecessary_blocking_io(options.avoid_unnecessary_blocking_io) {
}
void ImmutableDBOptions::Dump(Logger* log) const {
ROCKS_LOG_HEADER(log, " Options.error_if_exists: %d",
error_if_exists);
ROCKS_LOG_HEADER(log, " Options.create_if_missing: %d",
create_if_missing);
ROCKS_LOG_HEADER(log, " Options.paranoid_checks: %d",
paranoid_checks);
ROCKS_LOG_HEADER(log, " Options.env: %p",
env);
ROCKS_LOG_HEADER(log, " Options.info_log: %p",
info_log.get());
ROCKS_LOG_HEADER(log, " Options.max_file_opening_threads: %d",
max_file_opening_threads);
ROCKS_LOG_HEADER(log, " Options.statistics: %p",
statistics.get());
ROCKS_LOG_HEADER(log, " Options.use_fsync: %d",
use_fsync);
ROCKS_LOG_HEADER(
log, " Options.max_log_file_size: %" ROCKSDB_PRIszt,
max_log_file_size);
ROCKS_LOG_HEADER(log,
" Options.max_manifest_file_size: %" PRIu64,
max_manifest_file_size);
ROCKS_LOG_HEADER(
log, " Options.log_file_time_to_roll: %" ROCKSDB_PRIszt,
log_file_time_to_roll);
ROCKS_LOG_HEADER(
log, " Options.keep_log_file_num: %" ROCKSDB_PRIszt,
keep_log_file_num);
ROCKS_LOG_HEADER(
log, " Options.recycle_log_file_num: %" ROCKSDB_PRIszt,
recycle_log_file_num);
ROCKS_LOG_HEADER(log, " Options.allow_fallocate: %d",
allow_fallocate);
ROCKS_LOG_HEADER(log, " Options.allow_mmap_reads: %d",
allow_mmap_reads);
ROCKS_LOG_HEADER(log, " Options.allow_mmap_writes: %d",
allow_mmap_writes);
ROCKS_LOG_HEADER(log, " Options.use_direct_reads: %d",
use_direct_reads);
ROCKS_LOG_HEADER(log,
" "
"Options.use_direct_io_for_flush_and_compaction: %d",
use_direct_io_for_flush_and_compaction);
ROCKS_LOG_HEADER(log, " Options.create_missing_column_families: %d",
create_missing_column_families);
ROCKS_LOG_HEADER(log, " Options.db_log_dir: %s",
db_log_dir.c_str());
ROCKS_LOG_HEADER(log, " Options.wal_dir: %s",
wal_dir.c_str());
ROCKS_LOG_HEADER(log, " Options.table_cache_numshardbits: %d",
table_cache_numshardbits);
ROCKS_LOG_HEADER(log,
" Options.max_subcompactions: %" PRIu32,
max_subcompactions);
ROCKS_LOG_HEADER(log, " Options.max_background_flushes: %d",
max_background_flushes);
ROCKS_LOG_HEADER(log,
" Options.WAL_ttl_seconds: %" PRIu64,
wal_ttl_seconds);
ROCKS_LOG_HEADER(log,
" Options.WAL_size_limit_MB: %" PRIu64,
wal_size_limit_mb);
ROCKS_LOG_HEADER(
log, " Options.manifest_preallocation_size: %" ROCKSDB_PRIszt,
manifest_preallocation_size);
ROCKS_LOG_HEADER(log, " Options.is_fd_close_on_exec: %d",
is_fd_close_on_exec);
ROCKS_LOG_HEADER(log, " Options.advise_random_on_open: %d",
advise_random_on_open);
ROCKS_LOG_HEADER(
log, " Options.db_write_buffer_size: %" ROCKSDB_PRIszt,
db_write_buffer_size);
ROCKS_LOG_HEADER(log, " Options.write_buffer_manager: %p",
write_buffer_manager.get());
ROCKS_LOG_HEADER(log, " Options.access_hint_on_compaction_start: %d",
static_cast<int>(access_hint_on_compaction_start));
ROCKS_LOG_HEADER(log, " Options.new_table_reader_for_compaction_inputs: %d",
new_table_reader_for_compaction_inputs);
ROCKS_LOG_HEADER(
log, " Options.random_access_max_buffer_size: %" ROCKSDB_PRIszt,
random_access_max_buffer_size);
ROCKS_LOG_HEADER(log, " Options.use_adaptive_mutex: %d",
use_adaptive_mutex);
ROCKS_LOG_HEADER(log, " Options.rate_limiter: %p",
rate_limiter.get());
Header(
log, " Options.sst_file_manager.rate_bytes_per_sec: %" PRIi64,
sst_file_manager ? sst_file_manager->GetDeleteRateBytesPerSecond() : 0);
ROCKS_LOG_HEADER(log, " Options.wal_recovery_mode: %d",
static_cast<int>(wal_recovery_mode));
ROCKS_LOG_HEADER(log, " Options.enable_thread_tracking: %d",
enable_thread_tracking);
ROCKS_LOG_HEADER(log, " Options.enable_pipelined_write: %d",
enable_pipelined_write);
ROCKS_LOG_HEADER(log, " Options.allow_concurrent_memtable_write: %d",
allow_concurrent_memtable_write);
ROCKS_LOG_HEADER(log, " Options.enable_write_thread_adaptive_yield: %d",
enable_write_thread_adaptive_yield);
ROCKS_LOG_HEADER(log,
" Options.write_thread_max_yield_usec: %" PRIu64,
write_thread_max_yield_usec);
ROCKS_LOG_HEADER(log,
" Options.write_thread_slow_yield_usec: %" PRIu64,
write_thread_slow_yield_usec);
if (row_cache) {
ROCKS_LOG_HEADER(
log,
" Options.row_cache: %" ROCKSDB_PRIszt,
row_cache->GetCapacity());
} else {
ROCKS_LOG_HEADER(log,
" Options.row_cache: None");
}
#ifndef ROCKSDB_LITE
ROCKS_LOG_HEADER(log, " Options.wal_filter: %s",
wal_filter ? wal_filter->Name() : "None");
#endif // ROCKDB_LITE
ROCKS_LOG_HEADER(log, " Options.avoid_flush_during_recovery: %d",
avoid_flush_during_recovery);
ROCKS_LOG_HEADER(log, " Options.allow_ingest_behind: %d",
allow_ingest_behind);
Added support for differential snapshots 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
7 years ago
ROCKS_LOG_HEADER(log, " Options.preserve_deletes: %d",
preserve_deletes);
ROCKS_LOG_HEADER(log, " Options.two_write_queues: %d",
two_write_queues);
Optimize for serial commits in 2PC 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
7 years ago
ROCKS_LOG_HEADER(log, " Options.manual_wal_flush: %d",
manual_wal_flush);
ROCKS_LOG_HEADER(log, " Options.atomic_flush: %d", atomic_flush);
ROCKS_LOG_HEADER(log,
" Options.avoid_unnecessary_blocking_io: %d",
avoid_unnecessary_blocking_io);
}
MutableDBOptions::MutableDBOptions()
: max_background_jobs(2),
base_background_compactions(-1),
max_background_compactions(-1),
avoid_flush_during_shutdown(false),
writable_file_max_buffer_size(1024 * 1024),
delayed_write_rate(2 * 1024U * 1024U),
max_total_wal_size(0),
delete_obsolete_files_period_micros(6ULL * 60 * 60 * 1000000),
stats_dump_period_sec(600),
stats_persist_period_sec(600),
stats_history_buffer_size(1024 * 1024),
max_open_files(-1),
bytes_per_sync(0),
wal_bytes_per_sync(0),
compaction_readahead_size(0) {}
MutableDBOptions::MutableDBOptions(const DBOptions& options)
: max_background_jobs(options.max_background_jobs),
base_background_compactions(options.base_background_compactions),
max_background_compactions(options.max_background_compactions),
avoid_flush_during_shutdown(options.avoid_flush_during_shutdown),
writable_file_max_buffer_size(options.writable_file_max_buffer_size),
delayed_write_rate(options.delayed_write_rate),
max_total_wal_size(options.max_total_wal_size),
delete_obsolete_files_period_micros(
options.delete_obsolete_files_period_micros),
stats_dump_period_sec(options.stats_dump_period_sec),
stats_persist_period_sec(options.stats_persist_period_sec),
stats_history_buffer_size(options.stats_history_buffer_size),
max_open_files(options.max_open_files),
bytes_per_sync(options.bytes_per_sync),
wal_bytes_per_sync(options.wal_bytes_per_sync),
compaction_readahead_size(options.compaction_readahead_size) {}
void MutableDBOptions::Dump(Logger* log) const {
ROCKS_LOG_HEADER(log, " Options.max_background_jobs: %d",
max_background_jobs);
ROCKS_LOG_HEADER(log, " Options.max_background_compactions: %d",
max_background_compactions);
ROCKS_LOG_HEADER(log, " Options.avoid_flush_during_shutdown: %d",
avoid_flush_during_shutdown);
ROCKS_LOG_HEADER(
log, " Options.writable_file_max_buffer_size: %" ROCKSDB_PRIszt,
writable_file_max_buffer_size);
ROCKS_LOG_HEADER(log, " Options.delayed_write_rate : %" PRIu64,
delayed_write_rate);
ROCKS_LOG_HEADER(log, " Options.max_total_wal_size: %" PRIu64,
max_total_wal_size);
ROCKS_LOG_HEADER(
log, " Options.delete_obsolete_files_period_micros: %" PRIu64,
delete_obsolete_files_period_micros);
ROCKS_LOG_HEADER(log, " Options.stats_dump_period_sec: %u",
stats_dump_period_sec);
ROCKS_LOG_HEADER(log, " Options.stats_persist_period_sec: %d",
stats_persist_period_sec);
ROCKS_LOG_HEADER(
log,
" Options.stats_history_buffer_size: %" ROCKSDB_PRIszt,
stats_history_buffer_size);
ROCKS_LOG_HEADER(log, " Options.max_open_files: %d",
max_open_files);
ROCKS_LOG_HEADER(log,
" Options.bytes_per_sync: %" PRIu64,
bytes_per_sync);
ROCKS_LOG_HEADER(log,
" Options.wal_bytes_per_sync: %" PRIu64,
wal_bytes_per_sync);
ROCKS_LOG_HEADER(log,
" Options.compaction_readahead_size: %" ROCKSDB_PRIszt,
compaction_readahead_size);
}
} // namespace rocksdb