Revision 499ebb3ab5ea4207950fc95acf102b8f58add1c5 authored by Maysam Yabandeh on 24 June 2017, 21:06:43 UTC, committed by Facebook Github Bot on 24 June 2017, 21:11:29 UTC
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
1 parent 0ac4afb
Raw File
mutexlock.h
//  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.
//  This source code is also licensed under the GPLv2 license found in the
//  COPYING file in the root directory of this source tree.
//
// 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 <assert.h>
#include <atomic>
#include <mutex>
#include <thread>
#include "port/port.h"

namespace rocksdb {

// Helper class that locks a mutex on construction and unlocks the mutex when
// the destructor of the MutexLock object is invoked.
//
// Typical usage:
//
//   void MyClass::MyMethod() {
//     MutexLock l(&mu_);       // mu_ is an instance variable
//     ... some complex code, possibly with multiple return paths ...
//   }

class MutexLock {
 public:
  explicit MutexLock(port::Mutex *mu) : mu_(mu) {
    this->mu_->Lock();
  }
  ~MutexLock() { this->mu_->Unlock(); }

 private:
  port::Mutex *const mu_;
  // No copying allowed
  MutexLock(const MutexLock&);
  void operator=(const MutexLock&);
};

//
// Acquire a ReadLock on the specified RWMutex.
// The Lock will be automatically released then the
// object goes out of scope.
//
class ReadLock {
 public:
  explicit ReadLock(port::RWMutex *mu) : mu_(mu) {
    this->mu_->ReadLock();
  }
  ~ReadLock() { this->mu_->ReadUnlock(); }

 private:
  port::RWMutex *const mu_;
  // No copying allowed
  ReadLock(const ReadLock&);
  void operator=(const ReadLock&);
};

//
// Automatically unlock a locked mutex when the object is destroyed
//
class ReadUnlock {
 public:
  explicit ReadUnlock(port::RWMutex *mu) : mu_(mu) { mu->AssertHeld(); }
  ~ReadUnlock() { mu_->ReadUnlock(); }

 private:
  port::RWMutex *const mu_;
  // No copying allowed
  ReadUnlock(const ReadUnlock &) = delete;
  ReadUnlock &operator=(const ReadUnlock &) = delete;
};

//
// Acquire a WriteLock on the specified RWMutex.
// The Lock will be automatically released then the
// object goes out of scope.
//
class WriteLock {
 public:
  explicit WriteLock(port::RWMutex *mu) : mu_(mu) {
    this->mu_->WriteLock();
  }
  ~WriteLock() { this->mu_->WriteUnlock(); }

 private:
  port::RWMutex *const mu_;
  // No copying allowed
  WriteLock(const WriteLock&);
  void operator=(const WriteLock&);
};

//
// SpinMutex has very low overhead for low-contention cases.  Method names
// are chosen so you can use std::unique_lock or std::lock_guard with it.
//
class SpinMutex {
 public:
  SpinMutex() : locked_(false) {}

  bool try_lock() {
    auto currently_locked = locked_.load(std::memory_order_relaxed);
    return !currently_locked &&
           locked_.compare_exchange_weak(currently_locked, true,
                                         std::memory_order_acquire,
                                         std::memory_order_relaxed);
  }

  void lock() {
    for (size_t tries = 0;; ++tries) {
      if (try_lock()) {
        // success
        break;
      }
      port::AsmVolatilePause();
      if (tries > 100) {
        std::this_thread::yield();
      }
    }
  }

  void unlock() { locked_.store(false, std::memory_order_release); }

 private:
  std::atomic<bool> locked_;
};

}  // namespace rocksdb
back to top