Re: svn commit: r8935 - branches/1.0.x

On Tue, 2004-03-09 at 22:52, Greg Hudson wrote:
> On Tue, 2004-03-09 at 16:27, Sander Striker wrote:
> > Whether performance is acceptable is ofcourse something subjective
> > and rather specific to each site.
>
> Sometimes decision criteria are subjective.

*nods*

> > I personally think that reducing the number of times a file is opened
> > and parsed from the number of requests + subrequests (which can be
> > quite a lot) to one* can make a substantial difference.
>
> Perhaps. But CPUs are fast and disk reads are cached. Has anyone ever
> said, "I tried using mod_authz_svn but had to turn it off because
> Subversion got too slow with it on?" If so, I haven't seen it.

Not yet.

> > I guess I could go and produce numbers, but I don't really feel that's
> > worth the effort.
>
> There's nothing wrong with deciding that it isn't worth the effort, but
> that's not going to make me feel guilty.

No, no, that wasn't my intent anyway.

> The onus is on the person
> proposing the 1.0.x change to show that it meets the criteria for 1.0.x.

Which is fair.

Sander

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Mar 9 23:07:28 2004