W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1996

Re: Sticky stuff.

From: Anselm Baird-Smith <abaird@w3.org>
Date: Fri, 9 Aug 1996 18:41:53 +0500
Message-Id: <9608092241.AA02390@www18.w3.org>
To: Koen Holtman <koen@win.tue.nl>
Cc: Paul Leach <paulle@microsoft.com>, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, hallam@etna.ai.mit.edu
X-Mailing-List: <http-wg@cuckoo.hpl.hp.com> archive/latest/1282

 > >I think a new study would be needed to take these effects into account
 > >before we can conclude that sticky headers aren't worth the effrort.
 > I agree.  To get something like a firm conclusion, at least one other
 > study is needed.  My study was done a year ago, with a small sample
 > (145 Mb of traffic), and by someone who is not a statistician.
 > However, I think there is enough data to conclude that sticky headers
 > are *unlikely* to be worth the effort.  I therefore see no reason for
 > sticky headers to become a WG activity at this point.

As a server implementer, I feel that one thing is missing from this
dicussion: the (CPU) time to parse headers. Sticky header allows for
reusing parsed header values, and at least in my Java web server, this
account for a non-negligible portion of the CPU time that is used to
handle a request (by optimizing this part of the server I got up to
20% speed improvements). However, I don't know if this applies to C
server as well...

Of course, I beleive that - still in my case - the CPU time to handle
sticky headers would not be greater then the time to parse, say, a
real Accept header.

Received on Friday, 9 August 1996 15:43:33 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:40:17 UTC