为了解决BIO(同步阻塞IO)带来的一个链路需要一个线程处理的问题,后来有人对它的线程模型进行了优化,后端通过一个线程池来处理多个客户端的接入,从而形成了一个客户端个数M:线程池最大数N的关系。M可以远大于N,通过线程池可以灵活的调整线程资源,设置线程的最大值,防止由于大量并发导致资源耗尽。
服务端代码:
/**
* @author j.tommy
* Created by j.tommy on 2017/11/11.
*/
public class TimeServer {
private ThreadFactory namedThreadFactory = new ThreadFactoryBuilder().setNameFormat("demo-pool-%d").build();
private ExecutorService es = new ThreadPoolExecutor(1,100,0L, TimeUnit.MICROSECONDS,new LinkedBlockingQueue<Runnable>(1024),namedThreadFactory,new ThreadPoolExecutor.AbortPolicy());
public static void main(String[] args) {
TimeServer timeServer = new TimeServer();
timeServer.start(8080);
}
private void start(int port) {
ServerSocket ss = null;
try {
ss = new ServerSocket(port);
System.out.println("TimeServer is running...");
while (true) {
Socket socket = ss.accept();
es.submit(new TimeServerHandler(socket));
}
} catch (IOException e) {
e.printStackTrace();
} finally {
if (ss != null) {
try {
ss.close();
} catch (IOException e) {
e.printStackTrace();
}
es.shutdownNow();
}
}
}
class TimeServerHandler implements Runnable {
private Socket socket = null;
public TimeServerHandler(Socket socket) {
this.socket = socket;
}
@Override
public void run() {
BufferedReader br = null;
PrintWriter pw = null;
try {
br = new BufferedReader(new InputStreamReader(socket.getInputStream()));
pw = new PrintWriter(new OutputStreamWriter(socket.getOutputStream()),true);
while (true) {
String input = br.readLine();
if (null == input) {
break;
}
System.out.println("接收到客户端请求:" + input);
if ("Server Time".equalsIgnoreCase(input)) {
pw.println((new Date()).toString());
}
}
} catch (IOException e) {
System.out.println("ip:" + socket.getInetAddress().getHostAddress() + " I/O异常");
}
finally {
if (br != null) {
try {
br.close();
} catch (IOException e) {
e.printStackTrace();
}
}
if (pw != null) {
pw.close();
}
if (socket != null) {
try {
socket.close();
} catch (IOException e) {
e.printStackTrace();
}
}
}
}
}
}
客户端代码:
/**
* @author j.tommy
* Created by j.tommy on 2017/11/11.
*/
public class TimeClient {
public static void main(String[] args) {
TimeClient tc = new TimeClient();
tc.connect("127.0.0.1",8080);
}
private void connect(String host,int port) {
Socket socket = null;
InputStream in = null;
OutputStream out = null;
try {
socket = new Socket(host,port);
in = socket.getInputStream();
out = socket.getOutputStream();
PrintWriter pw = new PrintWriter(new OutputStreamWriter(out),true);
BufferedReader br = new BufferedReader(new InputStreamReader(in));
System.out.println("请求服务器时间");
pw.println("Server Time");
String response = br.readLine();
System.out.println("response:" + response);
} catch (IOException e) {
e.printStackTrace();
}
finally {
if (socket != null) {
try {
socket.close();
} catch (IOException e) {
e.printStackTrace();
}
}
if (in != null) {
try {
in.close();
} catch (IOException e) {
e.printStackTrace();
}
}
if (out != null) {
try {
out.close();
} catch (IOException e) {
e.printStackTrace();
}
}
}
}
}
伪异步IO通过线程池+任务队列来实现,当有新的客户端接入时,将客户端封装成一个新的Task,交给线程池处理。由于线程池和任务队列的大小都是可控制的,所以不论有多少个客户端访问,都不会导致资源耗尽和宕机。
伪异步IO的弊端:
当对Socket的输入流进行读取时,它会一直阻塞,直到发生下面3种事件之一:
1.有数据可读;
2.可用数据已经读取完毕;
3.发生空指针或IO异常。
这就意味着,当对方发送请求或应答消息缓慢、或者网络传输较慢时,读取输入流一方的线程将被长时间阻塞。在此期间,其他接入消息只能在任务队列中排队。
参考《Netty权威指南》