[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: [TSVN] Performances of TSVN vs Subversion

From: Mark Phippard <MarkP_at_softlanding.com>
Date: 2004-10-06 14:54:38 CEST

news <news@sea.gmane.org> wrote on 10/06/2004 04:42:07 AM:

> There has been a certain amount of rumours around Subversion 1.1.0 poor
> performances.
> Even if svn teams claims to have modified neon config to set
> #define HAVE_SETSOCKOPT 1
> it doesn't seem to change behaviour.
> In detail, PROPFIND requests are split in 2 packets: HTTP 1.1 header and

> DAV request body.
> Using svn command line, ethereal shows that, after sending first packet,

> there is an ACK received by the server with 200msec delay.
> Using TSVN the two packets are sent one after the other without waiting
> the ACK that arrives soon after.
> The total "svn ls URL" triggers 10 PROPFIND so with the command line
> there is an elapsed time of, at least, 2 seconds.
> In my env, TSVN completes in 0.14 msec !!!
>
> Is there anything else to do to achieve such performances.

Did you download the most recent Subversion 1.1.0 Windows binaries? They
claim to have been compiled with HAVE_SETSOCKOPT. If so, then post a
message to the list. Perhaps he had a pre-built Neon library and it
didn't rebuild when he made the new binary?

Mark

_____________________________________________________________________________
Scanned for SoftLanding Systems, Inc. by IBM Email Security Management Services powered by MessageLabs.
_____________________________________________________________________________

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tortoisesvn.tigris.org
For additional commands, e-mail: dev-help@tortoisesvn.tigris.org
Received on Wed Oct 6 16:02:13 2004

This is an archived mail posted to the TortoiseSVN Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.