【Boost】boost库asio详解3——io_service作为work pool
2016-03-15 09:25
513 查看
无论如何使用，都能感觉到使用boost.asio实现服务器，不仅是一件非常轻松的事，而且代码很漂亮，逻辑也相当清晰，这点上很不同于ACE。
使用io_service作为处理工作的work pool，可以看到，就是通过io_service.post投递一个Handler到io_service的队列,Handler在这个io_service.run内部得到执行，有可能你会发现，io_services.dispatch的接口也和io_service.post一样，但不同的是它是直接调用而不是经过push到队列然后在io_services.run中执行，而在这个示例当中，显然我们需要把工作交到另一个线程去完成，这样才不会影响网络接收线程池的工作以达到高效率的接收数据，这种设计与前面的netsever其实相同，这就是典型的Half
Sync/Half Async。二者的区别就是netsever自己实现了工作队列，而不是直接使用io_service，这种设计实际上在win下是使用了iocp作为工作队列。
不过我更倾向于前一种设计，因为那样做，代码一切都在自己的掌握中，而io_service则是经过许多封装代码，并且本身设计只是用于处理网络完成事件的。
无论如何使用，都能感觉到使用boost.asio实现服务器，不仅是一件非常轻松的事，而且代码很漂亮，逻辑也相当清晰，这点上很不同于ACE。
[cpp] view
plain copy
print?
#include <stdio.h>
#include <cstdlib>
#include <iostream>
#include <boost/thread.hpp>
#include <boost/aligned_storage.hpp>
#include <boost/array.hpp>
#include <boost/bind.hpp>
#include <boost/enable_shared_from_this.hpp>
#include <boost/noncopyable.hpp>
#include <boost/shared_ptr.hpp>
#include <boost/asio.hpp>
using boost::asio::ip::tcp;
class handler_allocator
: private boost::noncopyable
{
public:
handler_allocator()
: in_use_(false)
{
}
void* allocate(std::size_t size)
{
if (!in_use_ && size < storage_.size)
{
in_use_ = true;
return storage_.address();
}
else
{
return ::operator new(size);
}
}
void deallocate(void* pointer)
{
if (pointer == storage_.address())
{
in_use_ = false;
}
else
{
::operator delete(pointer);
}
}
private:
// Storage space used for handler-based custom memory allocation.
boost::aligned_storage<1024> storage_;
// Whether the handler-based custom allocation storage has been used.
bool in_use_;
};
template <typename Handler>
class custom_alloc_handler
{
public:
custom_alloc_handler(handler_allocator& a, Handler h)
: allocator_(a),
handler_(h)
{
}
template <typename Arg1>
void operator()(Arg1 arg1)
{
handler_(arg1);
}
template <typename Arg1, typename Arg2>
void operator()(Arg1 arg1, Arg2 arg2)
{
handler_(arg1, arg2);
}
friend void* asio_handler_allocate(std::size_t size,
custom_alloc_handler<Handler>* this_handler)
{
return this_handler->allocator_.allocate(size);
}
friend void asio_handler_deallocate(void* pointer, std::size_t /*size*/,
custom_alloc_handler<Handler>* this_handler)
{
this_handler->allocator_.deallocate(pointer);
}
private:
handler_allocator& allocator_;
Handler handler_;
};
// Helper function to wrap a handler object to add custom allocation.
template <typename Handler>
inline custom_alloc_handler<Handler> make_custom_alloc_handler(
handler_allocator& a, Handler h)
{
return custom_alloc_handler<Handler>(a, h);
}
/// A pool of io_service objects.
class io_service_pool
: private boost::noncopyable
{
public:
/// Construct the io_service pool.
explicit io_service_pool(std::size_t pool_size) : next_io_service_(0)
{
if (pool_size == 0)
throw std::runtime_error("io_service_pool size is 0");
// Give all the io_services work to do so that their run() functions will not
// exit until they are explicitly stopped.
for (std::size_t i = 0; i < pool_size; ++i)
{
io_service_ptr io_service(new boost::asio::io_service);
work_ptr work(new boost::asio::io_service::work(*io_service));
io_services_.push_back(io_service);
work_.push_back(work);
}
}
// Run all io_service objects in the pool.
void run()
{
// Create a pool of threads to run all of the io_services.
std::vector<boost::shared_ptr<boost::thread> > threads;
for (std::size_t i = 0; i < io_services_.size(); ++i)
{
boost::shared_ptr<boost::thread> thread(new boost::thread(
boost::bind(&boost::asio::io_service::run, io_services_[i])));
threads.push_back(thread);
}
// Wait for all threads in the pool to exit.
for (std::size_t i = 0; i < threads.size(); ++i)
threads[i]->join();
}
// Stop all io_service objects in the pool.
void stop()
{
// Explicitly stop all io_services.
for (std::size_t i = 0; i < io_services_.size(); ++i)
io_services_[i]->stop();
}
// Get an io_service to use.
boost::asio::io_service& get_io_service()
{
// Use a round-robin scheme to choose the next io_service to use.
boost::asio::io_service& io_service = *io_services_[next_io_service_];
++next_io_service_;
if (next_io_service_ == io_services_.size())
next_io_service_ = 0;
return io_service;
}
private:
typedef boost::shared_ptr<boost::asio::io_service> io_service_ptr;
typedef boost::shared_ptr<boost::asio::io_service::work> work_ptr;
/// The pool of io_services.
std::vector<io_service_ptr> io_services_;
/// The work that keeps the io_services running.
std::vector<work_ptr> work_;
/// The next io_service to use for a connection.
std::size_t next_io_service_;
};
class session
: public boost::enable_shared_from_this<session>
{
public:
session(boost::asio::io_service& work_service
, boost::asio::io_service& io_service)
: socket_(io_service)
, io_work_service(work_service)
{
}
tcp::socket& socket()
{
return socket_;
}
void start()
{
socket_.async_read_some(boost::asio::buffer(data_),
make_custom_alloc_handler(allocator_,
boost::bind(&session::handle_read,
shared_from_this(),
boost::asio::placeholders::error,
boost::asio::placeholders::bytes_transferred)));
}
void handle_read(const boost::system::error_code& error,
size_t bytes_transferred)
{
if (!error)
{
boost::shared_ptr<std::vector<char> > buf(new std::vector<char>);
buf->resize(bytes_transferred);
std::copy(data_.begin(), data_.begin() + bytes_transferred, buf->begin());
io_work_service.post(boost::bind(&session::on_receive
, shared_from_this(), buf, bytes_transferred));
socket_.async_read_some(boost::asio::buffer(data_),
make_custom_alloc_handler(allocator_,
boost::bind(&session::handle_read,
shared_from_this(),
boost::asio::placeholders::error,
boost::asio::placeholders::bytes_transferred)));
}
}
void handle_write(const boost::system::error_code& error)
{
if (!error)
{
}
}
void on_receive(boost::shared_ptr<std::vector<char> > buffers
, size_t bytes_transferred)
{
char* data_stream = &(*buffers->begin());
// in here finish the work.
std::cout << "receive :" << bytes_transferred << " bytes." <<
"message :" << data_stream << std::endl;
}
private:
// The io_service used to finish the work.
boost::asio::io_service& io_work_service;
// The socket used to communicate with the client.
tcp::socket socket_;
// Buffer used to store data received from the client.
boost::array<char, 1024> data_;
// The allocator to use for handler-based custom memory allocation.
handler_allocator allocator_;
};
typedef boost::shared_ptr<session> session_ptr;
class server
{
public:
server(short port, std::size_t io_service_pool_size)
: io_service_pool_(io_service_pool_size)
, io_service_work_pool_(io_service_pool_size)
, acceptor_(io_service_pool_.get_io_service(), tcp::endpoint(tcp::v4(), port))
{
session_ptr new_session(new session(io_service_work_pool_.get_io_service()
, io_service_pool_.get_io_service()));
acceptor_.async_accept(new_session->socket(),
boost::bind(&server::handle_accept, this, new_session,
boost::asio::placeholders::error));
}
void handle_accept(session_ptr new_session,
const boost::system::error_code& error)
{
if (!error)
{
new_session->start();
new_session.reset(new session(io_service_work_pool_.get_io_service()
, io_service_pool_.get_io_service()));
acceptor_.async_accept(new_session->socket(),
boost::bind(&server::handle_accept, this, new_session,
boost::asio::placeholders::error));
}
}
void run()
{
io_thread_.reset(new boost::thread(boost::bind(&io_service_pool::run
, &io_service_pool_)));
work_thread_.reset(new boost::thread(boost::bind(&io_service_pool::run
, &io_service_work_pool_)));
}
void stop()
{
io_service_pool_.stop();
io_service_work_pool_.stop();
io_thread_->join();
work_thread_->join();
}
private:
boost::shared_ptr<boost::thread> io_thread_;
boost::shared_ptr<boost::thread> work_thread_;
io_service_pool io_service_pool_;
io_service_pool io_service_work_pool_;
tcp::acceptor acceptor_;
};
int main(int argc, char* argv[])
{
try
{
if (argc != 2)
{
std::cerr << "Usage: server <port>/n";
return 1;
}
using namespace std; // For atoi.
server s(atoi(argv[1]), 10);
s.run();
getchar();
s.stop();
}
catch (std::exception& e)
{
std::cerr << "Exception: " << e.what() << "/n";
}
return 0;
}
使用io_service作为处理工作的work pool，可以看到，就是通过io_service.post投递一个Handler到io_service的队列,Handler在这个io_service.run内部得到执行，有可能你会发现，io_services.dispatch的接口也和io_service.post一样，但不同的是它是直接调用而不是经过push到队列然后在io_services.run中执行，而在这个示例当中，显然我们需要把工作交到另一个线程去完成，这样才不会影响网络接收线程池的工作以达到高效率的接收数据，这种设计与前面的netsever其实相同，这就是典型的Half
Sync/Half Async。二者的区别就是netsever自己实现了工作队列，而不是直接使用io_service，这种设计实际上在win下是使用了iocp作为工作队列。
不过我更倾向于前一种设计，因为那样做，代码一切都在自己的掌握中，而io_service则是经过许多封装代码，并且本身设计只是用于处理网络完成事件的。
无论如何使用，都能感觉到使用boost.asio实现服务器，不仅是一件非常轻松的事，而且代码很漂亮，逻辑也相当清晰，这点上很不同于ACE。
[cpp] view
plain copy
print?
#include <stdio.h>
#include <cstdlib>
#include <iostream>
#include <boost/thread.hpp>
#include <boost/aligned_storage.hpp>
#include <boost/array.hpp>
#include <boost/bind.hpp>
#include <boost/enable_shared_from_this.hpp>
#include <boost/noncopyable.hpp>
#include <boost/shared_ptr.hpp>
#include <boost/asio.hpp>
using boost::asio::ip::tcp;
class handler_allocator
: private boost::noncopyable
{
public:
handler_allocator()
: in_use_(false)
{
}
void* allocate(std::size_t size)
{
if (!in_use_ && size < storage_.size)
{
in_use_ = true;
return storage_.address();
}
else
{
return ::operator new(size);
}
}
void deallocate(void* pointer)
{
if (pointer == storage_.address())
{
in_use_ = false;
}
else
{
::operator delete(pointer);
}
}
private:
// Storage space used for handler-based custom memory allocation.
boost::aligned_storage<1024> storage_;
// Whether the handler-based custom allocation storage has been used.
bool in_use_;
};
template <typename Handler>
class custom_alloc_handler
{
public:
custom_alloc_handler(handler_allocator& a, Handler h)
: allocator_(a),
handler_(h)
{
}
template <typename Arg1>
void operator()(Arg1 arg1)
{
handler_(arg1);
}
template <typename Arg1, typename Arg2>
void operator()(Arg1 arg1, Arg2 arg2)
{
handler_(arg1, arg2);
}
friend void* asio_handler_allocate(std::size_t size,
custom_alloc_handler<Handler>* this_handler)
{
return this_handler->allocator_.allocate(size);
}
friend void asio_handler_deallocate(void* pointer, std::size_t /*size*/,
custom_alloc_handler<Handler>* this_handler)
{
this_handler->allocator_.deallocate(pointer);
}
private:
handler_allocator& allocator_;
Handler handler_;
};
// Helper function to wrap a handler object to add custom allocation.
template <typename Handler>
inline custom_alloc_handler<Handler> make_custom_alloc_handler(
handler_allocator& a, Handler h)
{
return custom_alloc_handler<Handler>(a, h);
}
/// A pool of io_service objects.
class io_service_pool
: private boost::noncopyable
{
public:
/// Construct the io_service pool.
explicit io_service_pool(std::size_t pool_size) : next_io_service_(0)
{
if (pool_size == 0)
throw std::runtime_error("io_service_pool size is 0");
// Give all the io_services work to do so that their run() functions will not
// exit until they are explicitly stopped.
for (std::size_t i = 0; i < pool_size; ++i)
{
io_service_ptr io_service(new boost::asio::io_service);
work_ptr work(new boost::asio::io_service::work(*io_service));
io_services_.push_back(io_service);
work_.push_back(work);
}
}
// Run all io_service objects in the pool.
void run()
{
// Create a pool of threads to run all of the io_services.
std::vector<boost::shared_ptr<boost::thread> > threads;
for (std::size_t i = 0; i < io_services_.size(); ++i)
{
boost::shared_ptr<boost::thread> thread(new boost::thread(
boost::bind(&boost::asio::io_service::run, io_services_[i])));
threads.push_back(thread);
}
// Wait for all threads in the pool to exit.
for (std::size_t i = 0; i < threads.size(); ++i)
threads[i]->join();
}
// Stop all io_service objects in the pool.
void stop()
{
// Explicitly stop all io_services.
for (std::size_t i = 0; i < io_services_.size(); ++i)
io_services_[i]->stop();
}
// Get an io_service to use.
boost::asio::io_service& get_io_service()
{
// Use a round-robin scheme to choose the next io_service to use.
boost::asio::io_service& io_service = *io_services_[next_io_service_];
++next_io_service_;
if (next_io_service_ == io_services_.size())
next_io_service_ = 0;
return io_service;
}
private:
typedef boost::shared_ptr<boost::asio::io_service> io_service_ptr;
typedef boost::shared_ptr<boost::asio::io_service::work> work_ptr;
/// The pool of io_services.
std::vector<io_service_ptr> io_services_;
/// The work that keeps the io_services running.
std::vector<work_ptr> work_;
/// The next io_service to use for a connection.
std::size_t next_io_service_;
};
class session
: public boost::enable_shared_from_this<session>
{
public:
session(boost::asio::io_service& work_service
, boost::asio::io_service& io_service)
: socket_(io_service)
, io_work_service(work_service)
{
}
tcp::socket& socket()
{
return socket_;
}
void start()
{
socket_.async_read_some(boost::asio::buffer(data_),
make_custom_alloc_handler(allocator_,
boost::bind(&session::handle_read,
shared_from_this(),
boost::asio::placeholders::error,
boost::asio::placeholders::bytes_transferred)));
}
void handle_read(const boost::system::error_code& error,
size_t bytes_transferred)
{
if (!error)
{
boost::shared_ptr<std::vector<char> > buf(new std::vector<char>);
buf->resize(bytes_transferred);
std::copy(data_.begin(), data_.begin() + bytes_transferred, buf->begin());
io_work_service.post(boost::bind(&session::on_receive
, shared_from_this(), buf, bytes_transferred));
socket_.async_read_some(boost::asio::buffer(data_),
make_custom_alloc_handler(allocator_,
boost::bind(&session::handle_read,
shared_from_this(),
boost::asio::placeholders::error,
boost::asio::placeholders::bytes_transferred)));
}
}
void handle_write(const boost::system::error_code& error)
{
if (!error)
{
}
}
void on_receive(boost::shared_ptr<std::vector<char> > buffers
, size_t bytes_transferred)
{
char* data_stream = &(*buffers->begin());
// in here finish the work.
std::cout << "receive :" << bytes_transferred << " bytes." <<
"message :" << data_stream << std::endl;
}
private:
// The io_service used to finish the work.
boost::asio::io_service& io_work_service;
// The socket used to communicate with the client.
tcp::socket socket_;
// Buffer used to store data received from the client.
boost::array<char, 1024> data_;
// The allocator to use for handler-based custom memory allocation.
handler_allocator allocator_;
};
typedef boost::shared_ptr<session> session_ptr;
class server
{
public:
server(short port, std::size_t io_service_pool_size)
: io_service_pool_(io_service_pool_size)
, io_service_work_pool_(io_service_pool_size)
, acceptor_(io_service_pool_.get_io_service(), tcp::endpoint(tcp::v4(), port))
{
session_ptr new_session(new session(io_service_work_pool_.get_io_service()
, io_service_pool_.get_io_service()));
acceptor_.async_accept(new_session->socket(),
boost::bind(&server::handle_accept, this, new_session,
boost::asio::placeholders::error));
}
void handle_accept(session_ptr new_session,
const boost::system::error_code& error)
{
if (!error)
{
new_session->start();
new_session.reset(new session(io_service_work_pool_.get_io_service()
, io_service_pool_.get_io_service()));
acceptor_.async_accept(new_session->socket(),
boost::bind(&server::handle_accept, this, new_session,
boost::asio::placeholders::error));
}
}
void run()
{
io_thread_.reset(new boost::thread(boost::bind(&io_service_pool::run
, &io_service_pool_)));
work_thread_.reset(new boost::thread(boost::bind(&io_service_pool::run
, &io_service_work_pool_)));
}
void stop()
{
io_service_pool_.stop();
io_service_work_pool_.stop();
io_thread_->join();
work_thread_->join();
}
private:
boost::shared_ptr<boost::thread> io_thread_;
boost::shared_ptr<boost::thread> work_thread_;
io_service_pool io_service_pool_;
io_service_pool io_service_work_pool_;
tcp::acceptor acceptor_;
};
int main(int argc, char* argv[])
{
try
{
if (argc != 2)
{
std::cerr << "Usage: server <port>/n";
return 1;
}
using namespace std; // For atoi.
server s(atoi(argv[1]), 10);
s.run();
getchar();
s.stop();
}
catch (std::exception& e)
{
std::cerr << "Exception: " << e.what() << "/n";
}
return 0;
}
内容来自用户分享和网络整理，不保证内容的准确性，如有侵权内容，可联系管理员处理 
相关文章推荐
- 使用spring redisTemplate.getValueSerializer()遇到的问题
- git 在Linux下搭建git服务器
- 去掉 URL 中讨厌的 jsessionid
- Platform Builder 6.0 180天试用过期
- 什么样的服务器才能叫做大并发
- maven web 项目中启动报错java.lang.ClassNotFoundException
- HTTPS与SSL(二)
- python 多线程
- 面试小结
- [POJ] 1948 Triangular Pastures (DP)
- 关于MySQL数据导出导入方法
- MySQL数据导出导入的方法
- listview的优化
- Android开发之动画效果浅析(一)
- 类，对象
- C++中的explicit关键字
- kNN(K-Nearest Neighbor)最邻近规则分类
- HTTPS与SSL(一)
- 手势的3个混淆属性 cancelstouchesinview delaystouchesbegan delaystouchesended
- Shiro源码学习之一