11526252cc
Summary:
PinnableSlice
Summary:
Currently the point lookup values are copied to a string provided by the
user. This incures an extra memcpy cost. This patch allows doing point lookup
via a PinnableSlice which pins the source memory location (instead of
copying their content) and releases them after the content is consumed
by the user. The old API of Get(string) is translated to the new API
underneath.
Here is the summary for improvements:
value 100 byte: 1.8% regular, 1.2% merge values
value 1k byte: 11.5% regular, 7.5% merge values
value 10k byte: 26% regular, 29.9% merge values
The improvement for merge could be more if we extend this approach to
pin the merge output and delay the full merge operation until the user
actually needs it. We have put that for future work.
PS:
Sometimes we observe a small decrease in performance when switching from
t5452014 to this patch but with the old Get(string) API. The d
Closes https://github.com/facebook/rocksdb/pull/1756
Differential Revision: D4391738
Pulled By: maysamyabandeh
fbshipit-source-id: 6f3edd3
89 lines
3.1 KiB
C++
89 lines
3.1 KiB
C++
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
|
// This source code is licensed under the BSD-style license found in the
|
|
// LICENSE file in the root directory of this source tree. An additional grant
|
|
// of patent rights can be found in the PATENTS file in the same directory.
|
|
|
|
#pragma once
|
|
#include <string>
|
|
#include "db/merge_context.h"
|
|
#include "db/range_del_aggregator.h"
|
|
#include "rocksdb/env.h"
|
|
#include "rocksdb/types.h"
|
|
#include "table/block.h"
|
|
|
|
namespace rocksdb {
|
|
class MergeContext;
|
|
class PinnedIteratorsManager;
|
|
|
|
class GetContext {
|
|
public:
|
|
enum GetState {
|
|
kNotFound,
|
|
kFound,
|
|
kDeleted,
|
|
kCorrupt,
|
|
kMerge // saver contains the current merge result (the operands)
|
|
};
|
|
|
|
GetContext(const Comparator* ucmp, const MergeOperator* merge_operator,
|
|
Logger* logger, Statistics* statistics, GetState init_state,
|
|
const Slice& user_key, PinnableSlice* value, bool* value_found,
|
|
MergeContext* merge_context, RangeDelAggregator* range_del_agg,
|
|
Env* env, SequenceNumber* seq = nullptr,
|
|
PinnedIteratorsManager* _pinned_iters_mgr = nullptr);
|
|
|
|
void MarkKeyMayExist();
|
|
|
|
// Records this key, value, and any meta-data (such as sequence number and
|
|
// state) into this GetContext.
|
|
//
|
|
// Returns True if more keys need to be read (due to merges) or
|
|
// False if the complete value has been found.
|
|
bool SaveValue(const ParsedInternalKey& parsed_key, const Slice& value,
|
|
Cleanable* value_pinner = nullptr);
|
|
|
|
// Simplified version of the previous function. Should only be used when we
|
|
// know that the operation is a Put.
|
|
void SaveValue(const Slice& value, SequenceNumber seq);
|
|
|
|
GetState State() const { return state_; }
|
|
|
|
RangeDelAggregator* range_del_agg() { return range_del_agg_; }
|
|
|
|
PinnedIteratorsManager* pinned_iters_mgr() { return pinned_iters_mgr_; }
|
|
|
|
// If a non-null string is passed, all the SaveValue calls will be
|
|
// logged into the string. The operations can then be replayed on
|
|
// another GetContext with replayGetContextLog.
|
|
void SetReplayLog(std::string* replay_log) { replay_log_ = replay_log; }
|
|
|
|
// Do we need to fetch the SequenceNumber for this key?
|
|
bool NeedToReadSequence() const { return (seq_ != nullptr); }
|
|
|
|
private:
|
|
const Comparator* ucmp_;
|
|
const MergeOperator* merge_operator_;
|
|
// the merge operations encountered;
|
|
Logger* logger_;
|
|
Statistics* statistics_;
|
|
|
|
GetState state_;
|
|
Slice user_key_;
|
|
PinnableSlice* pinnable_val_;
|
|
bool* value_found_; // Is value set correctly? Used by KeyMayExist
|
|
MergeContext* merge_context_;
|
|
RangeDelAggregator* range_del_agg_;
|
|
Env* env_;
|
|
// If a key is found, seq_ will be set to the SequenceNumber of most recent
|
|
// write to the key or kMaxSequenceNumber if unknown
|
|
SequenceNumber* seq_;
|
|
std::string* replay_log_;
|
|
// Used to temporarily pin blocks when state_ == GetContext::kMerge
|
|
PinnedIteratorsManager* pinned_iters_mgr_;
|
|
};
|
|
|
|
void replayGetContextLog(const Slice& replay_log, const Slice& user_key,
|
|
GetContext* get_context);
|
|
|
|
} // namespace rocksdb
|