RE: [Zope] Re: Zope performance ?
Chris wrote:
Given a "snapshot" of the Zope site, you could run webstones (from SGI) against it with a reasonable subset of pages, but again, it's all hand-waving in it's relation to reality---but that's the joy of benchmarks, they're only useful for proving that it's fast doing THAT.
Last time I looked at that benchmark (about 2 years ago), installing and using it was about 50 times harder than installing and using our server. [snip]
increases. mod_pcgi is probably the best option for glueing into Apache, in my opinion, and probably would reduce the problems 10 fold with authentication, etc.
OK, when can we have it? :^) Of course, while it reduces the current set of problems ten fold, it might just well create the same amount of new problems. --Paul Paul Everitt Digital Creations paul@digicool.com 540.371.6909
On Thu, Feb 25, 1999 at 09:23:19AM -0500, Paul Everitt wrote:
Chris wrote:
Given a "snapshot" of the Zope site, you could run webstones (from SGI) against it with a reasonable subset of pages, but again, it's all hand-waving in it's relation to reality---but that's the joy of benchmarks, they're only useful for proving that it's fast doing THAT.
Last time I looked at that benchmark (about 2 years ago), installing and using it was about 50 times harder than installing and using our server.
That sounds about right :-) I didn't say it was easy did I? :-) Actually, you could use async sockets in reverse (as a client solution) to write this in Python... maybe I'll work on that and include an example in the Library reference section I'm working on. Sadly, benchmarking is about 50% design, 30% setup, 1% test, 19% analysis... the test is easy, figuring out WHAT to test is the much harder part.
[snip]
increases. mod_pcgi is probably the best option for glueing into Apache, in my opinion, and probably would reduce the problems 10 fold with authentication, etc.
OK, when can we have it? :^) Of course, while it reduces the current set of problems ten fold, it might just well create the same amount of new problems.
Since I don't use Apache, I wouldn't even KNOW where to start ;-) You're right in that it could create some new problems, the big ones *I* see being: * Support of releases of Apache (1.2, 1.3, 1.4 soon) * Compile support Honestly, I'd say... "We support mod_pcgi on Apache 1.3.4 only, anything else and you're on your own, or use the CGI implementation." Most sites that are that concerned with performance, are also concerned with security, and so they track the Apache releases pretty religiously. Of late, from what I've heard from my company's developers, the Apache API has been relatively stable :-) Chris -- | Christopher Petrilli ``Television is bubble-gum for | petrilli@amber.org the mind.''-Frank Lloyd Wright
participants (2)
-
Christopher G. Petrilli -
Paul Everitt