HTTP管线化(HTTP pipelining)


    HTTP是一个简单的协议。客户进程建立一条同服务器进程的 T C P连接,然后发出请求并读取服务器进程的响应。服务器进程关闭连接表示本次响应结束。服务器进程返回的文件通常含有指向其他服务器上文件的指针(超文本链接)。用户显然可以很轻松地沿着这些链接从一个服务器到下一个服务器。 


    HTTP管线化是将多个HTTP要求(request)整批提交的技术,而在传送过程中不需先等待服务端的回应。管线化机制须通过永久连接(persistent connection)完成,仅HTTP/1.1支持此技术(HTTP/1.0不支持),并且只有GET和HEAD要求可以进行管线化,而POST则有所限制。此外,初次创建连接时也不应启动管线机制,因为对方(服务器)不一定支持HTTP/1.1版本的协议。



    在永久连接或者HTTP pipelining出现之前,每个连接的获取都需要创建一个独立的TCP连接。

    Prior to persistent connections, a separate TCP connection was established to fetch each URL, increasing the load on HTTP servers and causing congestion(拥塞) on the Internet.


    HTTP pipelining可以吧多个http请求输出到一个socket连接(倘若是基于socket的话,更确切的说应该是输出到一个传输层连接,在tcp ip中即为TCP连接)中去,然后等到对应的响应。

    HTTP pipelining is a technique in which multiple HTTP requests are written out to a single socket without waiting for the corresponding responses. Pipelining is only supported in HTTP/1.1, not in 1.0.
    The pipelining of requests results in a dramatic improvement in page loading times, especially over high latency connections such as satellite Internet connections.
    Since it is usually possible to fit several HTTP requests in the same TCP packet, HTTP pipelining allows fewer TCP packets to be sent over the network, reducing network load.
    Non-idempotent(非幂等) methods like POST should not be pipelined. Sequences of GET and HEAD requests can be always pipelined. A sequence of other idempotent requests like GET, HEAD, PUT and DELETE can be pipelined or not depending on whether requests in the sequence depend on the effect of others.[1]
    HTTP pipelining requires both the client and the server to support it. HTTP/1.1 conforming servers are required to support pipelining. This does not mean that servers are required to pipeline responses, but that they are required not to fail if a client chooses to pipeline requests.

Implementation in web servers 

    Implementing pipelining in web servers is a relatively simple matter of making sure that network buffers are not discarded between requests. For that reason, most modern web servers handle pipelining without any problem.
    Exceptions include IIS 4.

Implementation in web browsers 

    HTTP pipelining is disabled in most browsers.
    Opera has pipelining enabled by default. It uses heuristics to control the level of pipelining employed depending on the connected server.
    Internet Explorer 8 does not pipeline requests, due to concerns regarding buggy proxies and head-of-line blocking.
    Mozilla browsers (such as Mozilla Firefox, SeaMonkey and Camino), support pipelining however it is disabled by default.It uses some heuristics(测试试探), especially to turn pipelining off for IIS servers. does the same thing as Firefox.
    Konqueror 2.0 supports pipelining, but it's disabled by default.
    Google Chrome does not support pipelining.

Implementation in web proxies 
    Most HTTP proxies do not pipeline outgoing requests.
    Some versions of the Squid web proxy will pipeline up to two outgoing requests. This functionality has been disabled by default and needs to be manually enabled for "bandwidth management and access logging reasons."Squid supports multiple requests from clients.
    The Polipo proxy pipelines outgoing requests.