Re: Pipelining and compression effect on HTTP/1.1 proxies

> Luigi Rizzo:
> [...]
> >I have done a quick test on the content of our proxy cache: for each
> >directory, I have compared the output of
> [...]
> >with a saving, due to compression, of approximately 13% .
> 
> The 13% may be too pessimistic.
> 
> Another data point: about 2 years ago, I measured the amount of text/*
> data in the HTTP traffic between our campus proxy cache and outside
> servers, this turned out to be 30%.  As text/* data generally
> compresses with a factor of 75%, compression would lead to savings of
> 30%*0.75 = 23% for total off-campus traffic.

unfortunately WEB statistics are a moving target. Web pages nowadays
are much different than they were two years ago. We have moved from
1-2 inline graphics to the dozen that you can find on many such
pages.

> Also, compressing HTML pages may increase the perceived speed by more
> than this percentage because it would allow browsers to render the
> layout and start loading inlines (or subframes) earlier.

well, consider that modems also compress data, so the savings are not
as large as one could think. And, going along these lines, one would
then move to a binary HTTP and binary tags for HTML...

personally my feeling is that compression is an orthogonal problem,
which is already addressed at many layers (network; file system;
source encoding).  A content provider has all the interest to
provide compressed data.  If he chose not so, probably he'll pay
the consequence in the long term (because of user dissatisfaction),
and it is not the proxy code which has to take care of this.

> By the way, we really need more datapoints on this, and on other
> proposed optimisations.

Agreed.

	Cheers
	Luigi
-----------------------------+--------------------------------------
Luigi Rizzo                  |  Dip. di Ingegneria dell'Informazione
email: luigi@iet.unipi.it    |  Universita' di Pisa
tel: +39-50-568533           |  via Diotisalvi 2, 56126 PISA (Italy)
fax: +39-50-568522           |  http://www.iet.unipi.it/~luigi/
_____________________________|______________________________________

Received on Tuesday, 22 April 1997 11:57:24 UTC