RE: HTTP 2.0 and a Faster, more Mobile-friendly web

We have done a bit of work on looking at how SPDY works for apps (mostly simple trace analysis), but we don’t have anything significant to report yet. But this is indeed something that needs looking into.

From: James M Snell [mailto:jasnell@gmail.com]
Sent: Tuesday, July 31, 2012 1:00 PM
To: Henrik Frystyk Nielsen
Cc: ietf-http-wg@w3.org; Gabriel Montenegro; Rob Trace; Adalberto Foresti (MS OPEN TECH)
Subject: Re: HTTP 2.0 and a Faster, more Mobile-friendly web

Excellent to see this kind of analysis being performed. A few comments inline..
On Sun, Jul 29, 2012 at 1:14 PM, Henrik Frystyk Nielsen <henrikn@microsoft.com<mailto:henrikn@microsoft.com>> wrote:
[snip]

The SPDY proposal has been great for raising awareness of Web performance. It takes a “clean slate” approach to improving HTTP.

To compare the performance of SPDY with HTTP/1.1 we have run tests comparing download times of several public web sites using a controlled tested study. The test uses publically available software run with mostly default configurations while applying all the currently available optimizations to HTTP/1.1. You can find a preliminary report on the test results here: http://research.microsoft.com/apps/pubs/?id=170059. The results mirror other data (http://www.guypo.com/technical/not-as-spdy-as-you-thought) that indicate mixed results with SPDY performance.

Our results indicate almost equal performance between SPDY and HTTP/1.1 when one applies all the known optimizations to HTTP/1.1. SPDY’s performance improvements are not consistent and significant. We will continue our testing, and we welcome others to publish their results so that HTTP/2.0 can choose the best changes and deliver the best possible performance and scalability improvements compared to HTTP/1.1.


One of the frustrating issues for me, so far, is that nearly all of the discussion around HTTP/2.0 to date has centered around the requirements for browser clients; given the ever growing use HTTP as an enabler for integration APIs and the ever increasing number of non-browser HTTP client applications, it would be very interesting to formally compare and analyze the performance of SPDY vs. current HTTP/1.1 for *API* clients and servers. In my informal testing, I have seen very little difference, if any, between using SPDY* and a client using persistent HTTP/1.1 connections with pipelined API requests.
  * with the current http mapping defined in the spdy spec

With a few tweaks here and there to the SPDY+HTTP mapping (e.g. the binary header encoding I had kicked around previously) I was able to squeeze a bit more out the SPDY option but the results were certainly not conclusive.

This, of course, was not a formal study and my methodology was far from rigorous. A more thorough analysis would be required.

- James