|
|
|
// 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).
|
|
|
|
//
|
|
|
|
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
|
|
|
|
// Use of this source code is governed by a BSD-style license that can be
|
|
|
|
// found in the LICENSE file. See the AUTHORS file for names of contributors.
|
|
|
|
|
|
|
|
#pragma once
|
|
|
|
#include <algorithm>
|
|
|
|
#include <set>
|
|
|
|
#include <utility>
|
|
|
|
#include <vector>
|
|
|
|
#include <string>
|
|
|
|
#include "rocksdb/cache.h"
|
|
|
|
#include "db/dbformat.h"
|
|
|
|
#include "util/arena.h"
|
|
|
|
#include "util/autovector.h"
|
|
|
|
|
|
|
|
namespace rocksdb {
|
|
|
|
|
|
|
|
class VersionSet;
|
|
|
|
|
|
|
|
const uint64_t kFileNumberMask = 0x3FFFFFFFFFFFFFFF;
|
|
|
|
|
|
|
|
extern uint64_t PackFileNumberAndPathId(uint64_t number, uint64_t path_id);
|
|
|
|
|
|
|
|
// A copyable structure contains information needed to read data from an SST
|
|
|
|
// file. It can contain a pointer to a table reader opened for the file, or
|
|
|
|
// file number and size, which can be used to create a new table reader for it.
|
|
|
|
// The behavior is undefined when a copied of the structure is used when the
|
|
|
|
// file is not in any live version any more.
|
|
|
|
struct FileDescriptor {
|
|
|
|
// Table reader in table_reader_handle
|
|
|
|
TableReader* table_reader;
|
|
|
|
uint64_t packed_number_and_path_id;
|
|
|
|
uint64_t file_size; // File size in bytes
|
|
|
|
SequenceNumber smallest_seqno; // The smallest seqno in this file
|
|
|
|
SequenceNumber largest_seqno; // The largest seqno in this file
|
|
|
|
|
|
|
|
FileDescriptor() : FileDescriptor(0, 0, 0) {}
|
|
|
|
|
|
|
|
FileDescriptor(uint64_t number, uint32_t path_id, uint64_t _file_size)
|
|
|
|
: FileDescriptor(number, path_id, _file_size, kMaxSequenceNumber, 0) {}
|
|
|
|
|
|
|
|
FileDescriptor(uint64_t number, uint32_t path_id, uint64_t _file_size,
|
|
|
|
SequenceNumber _smallest_seqno, SequenceNumber _largest_seqno)
|
|
|
|
: table_reader(nullptr),
|
|
|
|
packed_number_and_path_id(PackFileNumberAndPathId(number, path_id)),
|
|
|
|
file_size(_file_size),
|
|
|
|
smallest_seqno(_smallest_seqno),
|
|
|
|
largest_seqno(_largest_seqno) {}
|
|
|
|
|
|
|
|
FileDescriptor& operator=(const FileDescriptor& fd) {
|
|
|
|
table_reader = fd.table_reader;
|
|
|
|
packed_number_and_path_id = fd.packed_number_and_path_id;
|
|
|
|
file_size = fd.file_size;
|
|
|
|
smallest_seqno = fd.smallest_seqno;
|
|
|
|
largest_seqno = fd.largest_seqno;
|
|
|
|
return *this;
|
|
|
|
}
|
|
|
|
|
|
|
|
uint64_t GetNumber() const {
|
|
|
|
return packed_number_and_path_id & kFileNumberMask;
|
|
|
|
}
|
|
|
|
uint32_t GetPathId() const {
|
|
|
|
return static_cast<uint32_t>(
|
|
|
|
packed_number_and_path_id / (kFileNumberMask + 1));
|
|
|
|
}
|
|
|
|
uint64_t GetFileSize() const { return file_size; }
|
|
|
|
};
|
|
|
|
|
|
|
|
struct FileSampledStats {
|
|
|
|
FileSampledStats() : num_reads_sampled(0) {}
|
|
|
|
FileSampledStats(const FileSampledStats& other) { *this = other; }
|
|
|
|
FileSampledStats& operator=(const FileSampledStats& other) {
|
|
|
|
num_reads_sampled = other.num_reads_sampled.load();
|
|
|
|
return *this;
|
|
|
|
}
|
|
|
|
|
|
|
|
// number of user reads to this file.
|
|
|
|
mutable std::atomic<uint64_t> num_reads_sampled;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct FileMetaData {
|
|
|
|
FileDescriptor fd;
|
|
|
|
InternalKey smallest; // Smallest internal key served by table
|
|
|
|
InternalKey largest; // Largest internal key served by table
|
|
|
|
|
|
|
|
// Needs to be disposed when refs becomes 0.
|
|
|
|
Cache::Handle* table_reader_handle;
|
|
|
|
|
|
|
|
FileSampledStats stats;
|
|
|
|
|
|
|
|
// Stats for compensating deletion entries during compaction
|
|
|
|
|
|
|
|
// File size compensated by deletion entry.
|
|
|
|
// This is updated in Version::UpdateAccumulatedStats() first time when the
|
|
|
|
// file is created or loaded. After it is updated (!= 0), it is immutable.
|
|
|
|
uint64_t compensated_file_size;
|
|
|
|
// These values can mutate, but they can only be read or written from
|
|
|
|
// single-threaded LogAndApply thread
|
|
|
|
uint64_t num_entries; // the number of entries.
|
|
|
|
uint64_t num_deletions; // the number of deletion entries.
|
|
|
|
uint64_t raw_key_size; // total uncompressed key size.
|
|
|
|
uint64_t raw_value_size; // total uncompressed value size.
|
|
|
|
|
|
|
|
int refs; // Reference count
|
|
|
|
|
|
|
|
bool being_compacted; // Is this file undergoing compaction?
|
|
|
|
bool init_stats_from_file; // true if the data-entry stats of this file
|
|
|
|
// has initialized from file.
|
|
|
|
|
Add experimental API MarkForCompaction()
Summary:
Some Mongo+Rocks datasets in Parse's environment are not doing compactions very frequently. During the quiet period (with no IO), we'd like to schedule compactions so that our reads become faster. Also, aggressively compacting during quiet periods helps when write bursts happen. In addition, we also want to compact files that are containing deleted key ranges (like old oplog keys).
All of this is currently not possible with CompactRange() because it's single-threaded and blocks all other compactions from happening. Running CompactRange() risks an issue of blocking writes because we generate too much Level 0 files before the compaction is over. Stopping writes is very dangerous because they hold transaction locks. We tried running manual compaction once on Mongo+Rocks and everything fell apart.
MarkForCompaction() solves all of those problems. This is very light-weight manual compaction. It is lower priority than automatic compactions, which means it shouldn't interfere with background process keeping the LSM tree clean. However, if no automatic compactions need to be run (or we have extra background threads available), we will start compacting files that are marked for compaction.
Test Plan: added a new unit test
Reviewers: yhchiang, rven, MarkCallaghan, sdong
Reviewed By: sdong
Subscribers: yoshinorim, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37083
10 years ago
|
|
|
bool marked_for_compaction; // True if client asked us nicely to compact this
|
|
|
|
// file.
|
|
|
|
|
|
|
|
FileMetaData()
|
|
|
|
: table_reader_handle(nullptr),
|
|
|
|
compensated_file_size(0),
|
|
|
|
num_entries(0),
|
|
|
|
num_deletions(0),
|
|
|
|
raw_key_size(0),
|
|
|
|
raw_value_size(0),
|
|
|
|
refs(0),
|
|
|
|
being_compacted(false),
|
Add experimental API MarkForCompaction()
Summary:
Some Mongo+Rocks datasets in Parse's environment are not doing compactions very frequently. During the quiet period (with no IO), we'd like to schedule compactions so that our reads become faster. Also, aggressively compacting during quiet periods helps when write bursts happen. In addition, we also want to compact files that are containing deleted key ranges (like old oplog keys).
All of this is currently not possible with CompactRange() because it's single-threaded and blocks all other compactions from happening. Running CompactRange() risks an issue of blocking writes because we generate too much Level 0 files before the compaction is over. Stopping writes is very dangerous because they hold transaction locks. We tried running manual compaction once on Mongo+Rocks and everything fell apart.
MarkForCompaction() solves all of those problems. This is very light-weight manual compaction. It is lower priority than automatic compactions, which means it shouldn't interfere with background process keeping the LSM tree clean. However, if no automatic compactions need to be run (or we have extra background threads available), we will start compacting files that are marked for compaction.
Test Plan: added a new unit test
Reviewers: yhchiang, rven, MarkCallaghan, sdong
Reviewed By: sdong
Subscribers: yoshinorim, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37083
10 years ago
|
|
|
init_stats_from_file(false),
|
|
|
|
marked_for_compaction(false) {}
|
|
|
|
|
|
|
|
// REQUIRED: Keys must be given to the function in sorted order (it expects
|
|
|
|
// the last key to be the largest).
|
|
|
|
void UpdateBoundaries(const Slice& key, SequenceNumber seqno) {
|
|
|
|
if (smallest.size() == 0) {
|
|
|
|
smallest.DecodeFrom(key);
|
|
|
|
}
|
|
|
|
largest.DecodeFrom(key);
|
|
|
|
fd.smallest_seqno = std::min(fd.smallest_seqno, seqno);
|
|
|
|
fd.largest_seqno = std::max(fd.largest_seqno, seqno);
|
|
|
|
}
|
Range deletion performance improvements + cleanup (#4014)
Summary:
This fixes the same performance issue that #3992 fixes but with much more invasive cleanup.
I'm more excited about this PR because it paves the way for fixing another problem we uncovered at Cockroach where range deletion tombstones can cause massive compactions. For example, suppose L4 contains deletions from [a, c) and [x, z) and no other keys, and L5 is entirely empty. L6, however, is full of data. When compacting L4 -> L5, we'll end up with one file that spans, massively, from [a, z). When we go to compact L5 -> L6, we'll have to rewrite all of L6! If, instead of range deletions in L4, we had keys a, b, x, y, and z, RocksDB would have been smart enough to create two files in L5: one for a and b and another for x, y, and z.
With the changes in this PR, it will be possible to adjust the compaction logic to split tombstones/start new output files when they would span too many files in the grandparent level.
ajkr please take a look when you have a minute!
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4014
Differential Revision: D8773253
Pulled By: ajkr
fbshipit-source-id: ec62fa85f648fdebe1380b83ed997f9baec35677
6 years ago
|
|
|
|
|
|
|
// Unlike UpdateBoundaries, ranges do not need to be presented in any
|
|
|
|
// particular order.
|
|
|
|
void UpdateBoundariesForRange(const InternalKey& start,
|
|
|
|
const InternalKey& end, SequenceNumber seqno,
|
|
|
|
const InternalKeyComparator& icmp) {
|
|
|
|
if (smallest.size() == 0 || icmp.Compare(start, smallest) < 0) {
|
|
|
|
smallest = start;
|
|
|
|
}
|
|
|
|
if (largest.size() == 0 || icmp.Compare(largest, end) < 0) {
|
|
|
|
largest = end;
|
|
|
|
}
|
|
|
|
fd.smallest_seqno = std::min(fd.smallest_seqno, seqno);
|
|
|
|
fd.largest_seqno = std::max(fd.largest_seqno, seqno);
|
Range deletion performance improvements + cleanup (#4014)
Summary:
This fixes the same performance issue that #3992 fixes but with much more invasive cleanup.
I'm more excited about this PR because it paves the way for fixing another problem we uncovered at Cockroach where range deletion tombstones can cause massive compactions. For example, suppose L4 contains deletions from [a, c) and [x, z) and no other keys, and L5 is entirely empty. L6, however, is full of data. When compacting L4 -> L5, we'll end up with one file that spans, massively, from [a, z). When we go to compact L5 -> L6, we'll have to rewrite all of L6! If, instead of range deletions in L4, we had keys a, b, x, y, and z, RocksDB would have been smart enough to create two files in L5: one for a and b and another for x, y, and z.
With the changes in this PR, it will be possible to adjust the compaction logic to split tombstones/start new output files when they would span too many files in the grandparent level.
ajkr please take a look when you have a minute!
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4014
Differential Revision: D8773253
Pulled By: ajkr
fbshipit-source-id: ec62fa85f648fdebe1380b83ed997f9baec35677
6 years ago
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
// A compressed copy of file meta data that just contain minimum data needed
|
|
|
|
// to server read operations, while still keeping the pointer to full metadata
|
|
|
|
// of the file in case it is needed.
|
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster
Summary:
Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena
Thus increase the file meta data locality, speed up "Get" and "FindFile"
benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing"
benchmark command:
./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920
Read Random:
From 1.8363 ms/op, improve to 1.7587 ms/op.
Read while writing:
From 2.985 ms/op, improve to 2.924 ms/op.
Test Plan:
make all check
Reviewers: ljin, haobo, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba, igor
Differential Revision: https://reviews.facebook.net/D19419
10 years ago
|
|
|
struct FdWithKeyRange {
|
|
|
|
FileDescriptor fd;
|
|
|
|
FileMetaData* file_metadata; // Point to all metadata
|
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster
Summary:
Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena
Thus increase the file meta data locality, speed up "Get" and "FindFile"
benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing"
benchmark command:
./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920
Read Random:
From 1.8363 ms/op, improve to 1.7587 ms/op.
Read while writing:
From 2.985 ms/op, improve to 2.924 ms/op.
Test Plan:
make all check
Reviewers: ljin, haobo, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba, igor
Differential Revision: https://reviews.facebook.net/D19419
10 years ago
|
|
|
Slice smallest_key; // slice that contain smallest key
|
|
|
|
Slice largest_key; // slice that contain largest key
|
|
|
|
|
|
|
|
FdWithKeyRange()
|
|
|
|
: fd(),
|
|
|
|
file_metadata(nullptr),
|
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster
Summary:
Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena
Thus increase the file meta data locality, speed up "Get" and "FindFile"
benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing"
benchmark command:
./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920
Read Random:
From 1.8363 ms/op, improve to 1.7587 ms/op.
Read while writing:
From 2.985 ms/op, improve to 2.924 ms/op.
Test Plan:
make all check
Reviewers: ljin, haobo, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba, igor
Differential Revision: https://reviews.facebook.net/D19419
10 years ago
|
|
|
smallest_key(),
|
|
|
|
largest_key() {
|
|
|
|
}
|
|
|
|
|
|
|
|
FdWithKeyRange(FileDescriptor _fd, Slice _smallest_key, Slice _largest_key,
|
|
|
|
FileMetaData* _file_metadata)
|
|
|
|
: fd(_fd),
|
|
|
|
file_metadata(_file_metadata),
|
|
|
|
smallest_key(_smallest_key),
|
|
|
|
largest_key(_largest_key) {}
|
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster
Summary:
Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena
Thus increase the file meta data locality, speed up "Get" and "FindFile"
benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing"
benchmark command:
./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920
Read Random:
From 1.8363 ms/op, improve to 1.7587 ms/op.
Read while writing:
From 2.985 ms/op, improve to 2.924 ms/op.
Test Plan:
make all check
Reviewers: ljin, haobo, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba, igor
Differential Revision: https://reviews.facebook.net/D19419
10 years ago
|
|
|
};
|
|
|
|
|
|
|
|
// Data structure to store an array of FdWithKeyRange in one level
|
|
|
|
// Actual data is guaranteed to be stored closely
|
|
|
|
struct LevelFilesBrief {
|
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster
Summary:
Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena
Thus increase the file meta data locality, speed up "Get" and "FindFile"
benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing"
benchmark command:
./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920
Read Random:
From 1.8363 ms/op, improve to 1.7587 ms/op.
Read while writing:
From 2.985 ms/op, improve to 2.924 ms/op.
Test Plan:
make all check
Reviewers: ljin, haobo, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba, igor
Differential Revision: https://reviews.facebook.net/D19419
10 years ago
|
|
|
size_t num_files;
|
|
|
|
FdWithKeyRange* files;
|
|
|
|
LevelFilesBrief() {
|
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster
Summary:
Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena
Thus increase the file meta data locality, speed up "Get" and "FindFile"
benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing"
benchmark command:
./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920
Read Random:
From 1.8363 ms/op, improve to 1.7587 ms/op.
Read while writing:
From 2.985 ms/op, improve to 2.924 ms/op.
Test Plan:
make all check
Reviewers: ljin, haobo, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba, igor
Differential Revision: https://reviews.facebook.net/D19419
10 years ago
|
|
|
num_files = 0;
|
|
|
|
files = nullptr;
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
class VersionEdit {
|
|
|
|
public:
|
|
|
|
VersionEdit() { Clear(); }
|
|
|
|
~VersionEdit() { }
|
|
|
|
|
|
|
|
void Clear();
|
|
|
|
|
|
|
|
void SetComparatorName(const Slice& name) {
|
|
|
|
has_comparator_ = true;
|
|
|
|
comparator_ = name.ToString();
|
|
|
|
}
|
|
|
|
void SetLogNumber(uint64_t num) {
|
|
|
|
has_log_number_ = true;
|
|
|
|
log_number_ = num;
|
|
|
|
}
|
|
|
|
void SetPrevLogNumber(uint64_t num) {
|
|
|
|
has_prev_log_number_ = true;
|
|
|
|
prev_log_number_ = num;
|
|
|
|
}
|
|
|
|
void SetNextFile(uint64_t num) {
|
|
|
|
has_next_file_number_ = true;
|
|
|
|
next_file_number_ = num;
|
|
|
|
}
|
|
|
|
void SetLastSequence(SequenceNumber seq) {
|
|
|
|
has_last_sequence_ = true;
|
|
|
|
last_sequence_ = seq;
|
|
|
|
}
|
|
|
|
void SetMaxColumnFamily(uint32_t max_column_family) {
|
|
|
|
has_max_column_family_ = true;
|
|
|
|
max_column_family_ = max_column_family;
|
|
|
|
}
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
7 years ago
|
|
|
void SetMinLogNumberToKeep(uint64_t num) {
|
|
|
|
has_min_log_number_to_keep_ = true;
|
|
|
|
min_log_number_to_keep_ = num;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool has_log_number() { return has_log_number_; }
|
|
|
|
|
|
|
|
uint64_t log_number() { return log_number_; }
|
|
|
|
|
|
|
|
// Add the specified file at the specified number.
|
|
|
|
// REQUIRES: This version has not been saved (see VersionSet::SaveTo)
|
|
|
|
// REQUIRES: "smallest" and "largest" are smallest and largest keys in file
|
|
|
|
void AddFile(int level, uint64_t file, uint32_t file_path_id,
|
|
|
|
uint64_t file_size, const InternalKey& smallest,
|
|
|
|
const InternalKey& largest, const SequenceNumber& smallest_seqno,
|
|
|
|
const SequenceNumber& largest_seqno,
|
|
|
|
bool marked_for_compaction) {
|
|
|
|
assert(smallest_seqno <= largest_seqno);
|
|
|
|
FileMetaData f;
|
|
|
|
f.fd = FileDescriptor(file, file_path_id, file_size, smallest_seqno,
|
|
|
|
largest_seqno);
|
|
|
|
f.smallest = smallest;
|
|
|
|
f.largest = largest;
|
|
|
|
f.fd.smallest_seqno = smallest_seqno;
|
|
|
|
f.fd.largest_seqno = largest_seqno;
|
|
|
|
f.marked_for_compaction = marked_for_compaction;
|
|
|
|
new_files_.emplace_back(level, std::move(f));
|
|
|
|
}
|
|
|
|
|
|
|
|
void AddFile(int level, const FileMetaData& f) {
|
|
|
|
assert(f.fd.smallest_seqno <= f.fd.largest_seqno);
|
|
|
|
new_files_.emplace_back(level, f);
|
|
|
|
}
|
|
|
|
|
|
|
|
// Delete the specified "file" from the specified "level".
|
|
|
|
void DeleteFile(int level, uint64_t file) {
|
|
|
|
deleted_files_.insert({level, file});
|
|
|
|
}
|
|
|
|
|
|
|
|
// Number of edits
|
|
|
|
size_t NumEntries() { return new_files_.size() + deleted_files_.size(); }
|
|
|
|
|
|
|
|
bool IsColumnFamilyManipulation() {
|
|
|
|
return is_column_family_add_ || is_column_family_drop_;
|
|
|
|
}
|
|
|
|
|
|
|
|
void SetColumnFamily(uint32_t column_family_id) {
|
|
|
|
column_family_ = column_family_id;
|
|
|
|
}
|
|
|
|
|
|
|
|
// set column family ID by calling SetColumnFamily()
|
|
|
|
void AddColumnFamily(const std::string& name) {
|
|
|
|
assert(!is_column_family_drop_);
|
|
|
|
assert(!is_column_family_add_);
|
|
|
|
assert(NumEntries() == 0);
|
|
|
|
is_column_family_add_ = true;
|
|
|
|
column_family_name_ = name;
|
|
|
|
}
|
|
|
|
|
|
|
|
// set column family ID by calling SetColumnFamily()
|
|
|
|
void DropColumnFamily() {
|
|
|
|
assert(!is_column_family_drop_);
|
|
|
|
assert(!is_column_family_add_);
|
|
|
|
assert(NumEntries() == 0);
|
|
|
|
is_column_family_drop_ = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
// return true on success.
|
|
|
|
bool EncodeTo(std::string* dst) const;
|
|
|
|
Status DecodeFrom(const Slice& src);
|
|
|
|
|
|
|
|
const char* DecodeNewFile4From(Slice* input);
|
|
|
|
|
|
|
|
typedef std::set<std::pair<int, uint64_t>> DeletedFileSet;
|
|
|
|
|
|
|
|
const DeletedFileSet& GetDeletedFiles() { return deleted_files_; }
|
|
|
|
const std::vector<std::pair<int, FileMetaData>>& GetNewFiles() {
|
|
|
|
return new_files_;
|
|
|
|
}
|
|
|
|
|
|
|
|
void MarkAtomicGroup(uint32_t remaining_entries) {
|
|
|
|
is_in_atomic_group_ = true;
|
|
|
|
remaining_entries_ = remaining_entries;
|
|
|
|
}
|
|
|
|
|
|
|
|
std::string DebugString(bool hex_key = false) const;
|
Added JSON manifest dump option to ldb command
Summary:
Added a new flag --json to the ldb manifest_dump command
that prints out the version edits as JSON objects for easier
reading and parsing of information.
Test Plan:
**Sample usage: **
```
./ldb manifest_dump --json --path=path/to/manifest/file
```
**Sample output:**
```
{"EditNumber": 0, "Comparator": "leveldb.BytewiseComparator", "ColumnFamily": 0}
{"EditNumber": 1, "LogNumber": 0, "ColumnFamily": 0}
{"EditNumber": 2, "LogNumber": 4, "PrevLogNumber": 0, "NextFileNumber": 7, "LastSeq": 35356, "AddedFiles": [{"Level": 0, "FileNumber": 5, "FileSize": 1949284, "SmallestIKey": "'", "LargestIKey": "'"}], "ColumnFamily": 0}
...
{"EditNumber": 13, "PrevLogNumber": 0, "NextFileNumber": 36, "LastSeq": 290994, "DeletedFiles": [{"Level": 0, "FileNumber": 17}, {"Level": 0, "FileNumber": 20}, {"Level": 0, "FileNumber": 22}, {"Level": 0, "FileNumber": 24}, {"Level": 1, "FileNumber": 13}, {"Level": 1, "FileNumber": 14}, {"Level": 1, "FileNumber": 15}, {"Level": 1, "FileNumber": 18}], "AddedFiles": [{"Level": 1, "FileNumber": 25, "FileSize": 2114340, "SmallestIKey": "'", "LargestIKey": "'"}, {"Level": 1, "FileNumber": 26, "FileSize": 2115213, "SmallestIKey": "'", "LargestIKey": "'"}, {"Level": 1, "FileNumber": 27, "FileSize": 2114807, "SmallestIKey": "'", "LargestIKey": "'"}, {"Level": 1, "FileNumber": 30, "FileSize": 2115271, "SmallestIKey": "'", "LargestIKey": "'"}, {"Level": 1, "FileNumber": 31, "FileSize": 2115165, "SmallestIKey": "'", "LargestIKey": "'"}, {"Level": 1, "FileNumber": 32, "FileSize": 2114683, "SmallestIKey": "'", "LargestIKey": "'"}, {"Level": 1, "FileNumber": 35, "FileSize": 1757512, "SmallestIKey": "'", "LargestIKey": "'"}], "ColumnFamily": 0}
...
```
Reviewers: sdong, anthony, yhchiang, igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D41727
9 years ago
|
|
|
std::string DebugJSON(int edit_num, bool hex_key = false) const;
|
|
|
|
|
|
|
|
private:
|
|
|
|
friend class VersionSet;
|
|
|
|
friend class Version;
|
|
|
|
|
|
|
|
bool GetLevel(Slice* input, int* level, const char** msg);
|
|
|
|
|
|
|
|
int max_level_;
|
|
|
|
std::string comparator_;
|
|
|
|
uint64_t log_number_;
|
|
|
|
uint64_t prev_log_number_;
|
|
|
|
uint64_t next_file_number_;
|
|
|
|
uint32_t max_column_family_;
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
7 years ago
|
|
|
// The most recent WAL log number that is deleted
|
|
|
|
uint64_t min_log_number_to_keep_;
|
|
|
|
SequenceNumber last_sequence_;
|
|
|
|
bool has_comparator_;
|
|
|
|
bool has_log_number_;
|
|
|
|
bool has_prev_log_number_;
|
|
|
|
bool has_next_file_number_;
|
|
|
|
bool has_last_sequence_;
|
|
|
|
bool has_max_column_family_;
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
7 years ago
|
|
|
bool has_min_log_number_to_keep_;
|
|
|
|
|
|
|
|
DeletedFileSet deleted_files_;
|
|
|
|
std::vector<std::pair<int, FileMetaData>> new_files_;
|
|
|
|
|
|
|
|
// Each version edit record should have column_family_ set
|
|
|
|
// If it's not set, it is default (0)
|
|
|
|
uint32_t column_family_;
|
|
|
|
// a version edit can be either column_family add or
|
|
|
|
// column_family drop. If it's column family add,
|
|
|
|
// it also includes column family name.
|
|
|
|
bool is_column_family_drop_;
|
|
|
|
bool is_column_family_add_;
|
|
|
|
std::string column_family_name_;
|
|
|
|
|
|
|
|
bool is_in_atomic_group_;
|
|
|
|
uint32_t remaining_entries_;
|
|
|
|
};
|
|
|
|
|
|
|
|
} // namespace rocksdb
|