3dff28cf9b
Summary: For performance purposes, the lower level routines were changed to use a SystemClock* instead of a std::shared_ptr<SystemClock>. The shared ptr has some performance degradation on certain hardware classes. For most of the system, there is no risk of the pointer being deleted/invalid because the shared_ptr will be stored elsewhere. For example, the ImmutableDBOptions stores the Env which has a std::shared_ptr<SystemClock> in it. The SystemClock* within the ImmutableDBOptions is essentially a "short cut" to gain access to this constant resource. There were a few classes (PeriodicWorkScheduler?) where the "short cut" property did not hold. In those cases, the shared pointer was preserved. Using db_bench readrandom perf_level=3 on my EC2 box, this change performed as well or better than 6.17: 6.17: readrandom : 28.046 micros/op 854902 ops/sec; 61.3 MB/s (355999 of 355999 found) 6.18: readrandom : 32.615 micros/op 735306 ops/sec; 52.7 MB/s (290999 of 290999 found) PR: readrandom : 27.500 micros/op 871909 ops/sec; 62.5 MB/s (367999 of 367999 found) (Note that the times for 6.18 are prior to revert of the SystemClock). Pull Request resolved: https://github.com/facebook/rocksdb/pull/8033 Reviewed By: pdillinger Differential Revision: D27014563 Pulled By: mrambacher fbshipit-source-id: ad0459eba03182e454391b5926bf5cdd45657b67
115 lines
3.4 KiB
C++
115 lines
3.4 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
|
|
|
|
#include <algorithm>
|
|
#include <atomic>
|
|
#include <chrono>
|
|
#include <deque>
|
|
|
|
#include "port/port.h"
|
|
#include "rocksdb/env.h"
|
|
#include "rocksdb/rate_limiter.h"
|
|
#include "rocksdb/system_clock.h"
|
|
#include "util/mutexlock.h"
|
|
#include "util/random.h"
|
|
|
|
namespace ROCKSDB_NAMESPACE {
|
|
|
|
class GenericRateLimiter : public RateLimiter {
|
|
public:
|
|
GenericRateLimiter(int64_t refill_bytes, int64_t refill_period_us,
|
|
int32_t fairness, RateLimiter::Mode mode,
|
|
const std::shared_ptr<SystemClock>& clock,
|
|
bool auto_tuned);
|
|
|
|
virtual ~GenericRateLimiter();
|
|
|
|
// This API allows user to dynamically change rate limiter's bytes per second.
|
|
virtual void SetBytesPerSecond(int64_t bytes_per_second) override;
|
|
|
|
// Request for token to write bytes. If this request can not be satisfied,
|
|
// the call is blocked. Caller is responsible to make sure
|
|
// bytes <= GetSingleBurstBytes()
|
|
using RateLimiter::Request;
|
|
virtual void Request(const int64_t bytes, const Env::IOPriority pri,
|
|
Statistics* stats) override;
|
|
|
|
virtual int64_t GetSingleBurstBytes() const override {
|
|
return refill_bytes_per_period_.load(std::memory_order_relaxed);
|
|
}
|
|
|
|
virtual int64_t GetTotalBytesThrough(
|
|
const Env::IOPriority pri = Env::IO_TOTAL) const override {
|
|
MutexLock g(&request_mutex_);
|
|
if (pri == Env::IO_TOTAL) {
|
|
return total_bytes_through_[Env::IO_LOW] +
|
|
total_bytes_through_[Env::IO_HIGH];
|
|
}
|
|
return total_bytes_through_[pri];
|
|
}
|
|
|
|
virtual int64_t GetTotalRequests(
|
|
const Env::IOPriority pri = Env::IO_TOTAL) const override {
|
|
MutexLock g(&request_mutex_);
|
|
if (pri == Env::IO_TOTAL) {
|
|
return total_requests_[Env::IO_LOW] + total_requests_[Env::IO_HIGH];
|
|
}
|
|
return total_requests_[pri];
|
|
}
|
|
|
|
virtual int64_t GetBytesPerSecond() const override {
|
|
return rate_bytes_per_sec_;
|
|
}
|
|
|
|
private:
|
|
void Refill();
|
|
int64_t CalculateRefillBytesPerPeriod(int64_t rate_bytes_per_sec);
|
|
Status Tune();
|
|
|
|
uint64_t NowMicrosMonotonic() { return clock_->NowNanos() / std::milli::den; }
|
|
|
|
// This mutex guard all internal states
|
|
mutable port::Mutex request_mutex_;
|
|
|
|
const int64_t kMinRefillBytesPerPeriod = 100;
|
|
|
|
const int64_t refill_period_us_;
|
|
|
|
int64_t rate_bytes_per_sec_;
|
|
// This variable can be changed dynamically.
|
|
std::atomic<int64_t> refill_bytes_per_period_;
|
|
std::shared_ptr<SystemClock> clock_;
|
|
|
|
bool stop_;
|
|
port::CondVar exit_cv_;
|
|
int32_t requests_to_wait_;
|
|
|
|
int64_t total_requests_[Env::IO_TOTAL];
|
|
int64_t total_bytes_through_[Env::IO_TOTAL];
|
|
int64_t available_bytes_;
|
|
int64_t next_refill_us_;
|
|
|
|
int32_t fairness_;
|
|
Random rnd_;
|
|
|
|
struct Req;
|
|
Req* leader_;
|
|
std::deque<Req*> queue_[Env::IO_TOTAL];
|
|
|
|
bool auto_tuned_;
|
|
int64_t num_drains_;
|
|
int64_t prev_num_drains_;
|
|
const int64_t max_bytes_per_sec_;
|
|
std::chrono::microseconds tuned_time_;
|
|
};
|
|
|
|
} // namespace ROCKSDB_NAMESPACE
|