rocksdb/db/compaction_picker_universal.h
Anand Ananthabhotla a736255de8 Delete triggered compaction for universal style
Summary:
This is still WIP, but I'm hoping for early feedback on the overall approach.

This patch implements deletion triggered compaction, which till now only
worked for leveled, for universal style. SST files are marked for
compaction by the CompactOnDeletionCollertor table property. This is
expected to be used when free disk space is low and the user wants to
reclaim space by deleting a bunch of keys. The deletions are expected to
be dense. In such a situation, we want to avoid a full compaction due to
its space overhead.

The strategy used in this case is similar to leveled. We pick one file
from the set of files marked for compaction. We then expand the inputs
to a clean cut on the same level, and then pick overlapping files from
the next non-mepty level. Picking files from the next level can cause
the key range to expand, and we opportunistically expand inputs in the
source level to include files wholly in this key range.

The main side effect of this is that it breaks the property of no time
range overlap between levels. This shouldn't break any functionality.
Closes https://github.com/facebook/rocksdb/pull/3860

Differential Revision: D8124397

Pulled By: anand1976

fbshipit-source-id: bfa2a9dd6817930e991b35d3a8e7e61304ed3dcf
2018-05-29 15:44:34 -07:00

99 lines
4 KiB
C++

// 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
#ifndef ROCKSDB_LITE
#include "db/compaction_picker.h"
namespace rocksdb {
class UniversalCompactionPicker : public CompactionPicker {
public:
UniversalCompactionPicker(const ImmutableCFOptions& ioptions,
const InternalKeyComparator* icmp)
: CompactionPicker(ioptions, icmp) {}
virtual Compaction* PickCompaction(const std::string& cf_name,
const MutableCFOptions& mutable_cf_options,
VersionStorageInfo* vstorage,
LogBuffer* log_buffer) override;
virtual int MaxOutputLevel() const override { return NumberLevels() - 1; }
virtual bool NeedsCompaction(
const VersionStorageInfo* vstorage) const override;
private:
struct SortedRun {
SortedRun(int _level, FileMetaData* _file, uint64_t _size,
uint64_t _compensated_file_size, bool _being_compacted)
: level(_level),
file(_file),
size(_size),
compensated_file_size(_compensated_file_size),
being_compacted(_being_compacted) {
assert(compensated_file_size > 0);
assert(level != 0 || file != nullptr);
}
void Dump(char* out_buf, size_t out_buf_size,
bool print_path = false) const;
// sorted_run_count is added into the string to print
void DumpSizeInfo(char* out_buf, size_t out_buf_size,
size_t sorted_run_count) const;
int level;
// `file` Will be null for level > 0. For level = 0, the sorted run is
// for this file.
FileMetaData* file;
// For level > 0, `size` and `compensated_file_size` are sum of sizes all
// files in the level. `being_compacted` should be the same for all files
// in a non-zero level. Use the value here.
uint64_t size;
uint64_t compensated_file_size;
bool being_compacted;
};
// Pick Universal compaction to limit read amplification
Compaction* PickCompactionToReduceSortedRuns(
const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
VersionStorageInfo* vstorage, double score, unsigned int ratio,
unsigned int num_files, const std::vector<SortedRun>& sorted_runs,
LogBuffer* log_buffer);
// Pick Universal compaction to limit space amplification.
Compaction* PickCompactionToReduceSizeAmp(
const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
VersionStorageInfo* vstorage, double score,
const std::vector<SortedRun>& sorted_runs, LogBuffer* log_buffer);
Compaction* PickDeleteTriggeredCompaction(
const std::string& cf_name, const MutableCFOptions& mutable_cf_options,
VersionStorageInfo* vstorage, double score,
const std::vector<SortedRun>& sorted_runs, LogBuffer* log_buffer);
// Used in universal compaction when the enabled_trivial_move
// option is set. Checks whether there are any overlapping files
// in the input. Returns true if the input files are non
// overlapping.
bool IsInputFilesNonOverlapping(Compaction* c);
static std::vector<SortedRun> CalculateSortedRuns(
const VersionStorageInfo& vstorage, const ImmutableCFOptions& ioptions,
const MutableCFOptions& mutable_cf_options);
// Pick a path ID to place a newly generated file, with its estimated file
// size.
static uint32_t GetPathId(const ImmutableCFOptions& ioptions,
const MutableCFOptions& mutable_cf_options,
uint64_t file_size);
};
} // namespace rocksdb
#endif // !ROCKSDB_LITE