0

I'm using a program to monitor the browser's activity, when I go to a site it makes an HTTP request over a TCP connection and I get the HTML file from the root directory of the server and then, it gets the image objects inside that HTML file but they're on a different directory and so it creates a new connection... Is this what HTTP/1.1 does when it is pipelining? Or is it doing this for some other reason? I've been reading the RFC 2616 and it doesn't mention why it creates these new connections.

besnico
  • 253
  • 2
  • 4
  • 13

1 Answers1

0

It is doing that simultaneously so it can download all the images, javascripts, CSS files, etc., in parallel. Hence the new connections.

user207421
  • 305,947
  • 44
  • 307
  • 483
  • is that a notion present on the RFC 2616? Because the page has 3 images, why didn't it open 4 connections if the maximum number of simultaneous connections is 4? – besnico Apr 16 '12 at 00:50
  • No, it isn't in the RFC, it is a very long-standing browser programming practice, so that for example all images start displaying at the same time rather than consecutively, which gives a better user experience. The browser will have some limit on the number of simultaneous connections. – user207421 Apr 16 '12 at 00:58