使用java计算Tcp往返时间 [英] Tcp round trip time calculation using java

问题描述

当我使用 10 gig 网络接口从一台 solaris m/c 跟踪路由到另一台时，40 字节数据包需要 0.073 毫秒.

When i do traceroute from one solaris m/c to another using 10 gig network interface it takes 0.073 ms for 40 bytes packets.

当我在 Java 中做同样的事情时，时间会更长.即使经过 10K 次迭代，它也更长.可能是什么原因 ?

When i do the same thing in java, the time is way longer. It is longer even after 10K iteration. What could be the reason ?

Java:发件人(代码段)

Java: Sender (Snippet)

Socket sendingSocket = new Socket(address, RECEIVER_PORT);
 sendingSocket.setTcpNoDelay(true);
 OutputStream outputStream = sendingSocket.getOutputStream();
 byte[] msg = new byte[64]; // assume that it is populated. 
 for (int i = 0; i < 10000; i++) {
    long start = System.nanoTime();
    outputStream.write(msg,0,64);
    outputStream.flush();

    inputStream.read(msg,0,64);  // inputStream is initialized like outputstream 
    long end = System.nanoTime(); 
 }  

它需要更长的时间 69 毫秒，它甚至不取决于字节大小.即使我将其减少为 1 个字节的数组，它仍然需要 69 毫秒.任何评论/建议?

It takes way longer 69 millis and it does not even depends upon the byte size. Even if i reduce it to say 1 byte array, it still takes 69 millis. Any comment/Suggestion ?

其他观察:1. OutputStream.write 和flush 只需要6 微秒.2.同样在另一端TCPReceiver端接收和写回，只需要6微秒.

Other Observation: 1. OutputStream.write and flush only takes 6 micros. 2. Similarly on the other end TCPReceiver side which receives and writes back, it only takes 6 micros.

解决方案:谢谢大家对这个问题的回复.我发现这是由于套接字缓冲区大小:

Solution: Thank you all you responded for this query. I found that it was due to the socket buffer size:

在 solaris m/c 上设置的默认缓冲区大小.

Default buffer size set on solaris m/c.

接收缓冲区大小 49152.

Received Buffer Size 49152.

发送缓冲区大小 7552.

Sending Buffer Size 7552.

我增加了套接字缓冲区大小，性能几乎与 traceRoute 匹配.

I increased the socket buffer size and the performance almost matches traceRoute.

推荐答案

你不是在比较喜欢和喜欢.ICMP 和 TCP 是不同的协议.

You are not comparing like with like. ICMP and TCP are way different protocols.

如果您想确定延迟是在您的代码、JVM、Solaris TCP 堆栈还是网络中，您必须从 tcpdump/wireshark 等开始.

If you want to decide if the latency is in your code, the JVM, Solaris TCP stack or the network you'll have to start with tcpdump / wireshark etc.

这篇关于使用java计算Tcp往返时间的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！