|
From: | Vincent Caron |
Subject: | Re: [Savannah-hackers] Savannah CVS update performance... |
Date: | Thu, 20 Nov 2003 02:38:30 +0100 |
User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031014 Thunderbird/0.3 |
Eric Blossom wrote:
I've attached a log below. Note that there's only a single modified file in the tree, tx_am.py and it's only 2KB long. I don't think your problem is bandwidth, looks like something else. Taking 1:20 to determine that there's a single file modified seems like about 1:10 too long...
The same operation with the same context takes 4 sec locally on the CVS server, does not looks like a load limitation.
It just took me a whoping 6 sec to do the same test from Paris, France. I'm on a regular home 512k DSL. Does not look like a bandwidth problem.
I'm not sure the load and bandwidth parameters have changed a lot since 1 hour when you ran your test (at least in that order of magnitude : 10x slower according to you). Sorry, but no evidence that Savannah is at fault :).
[Prev in Thread] | Current Thread | [Next in Thread] |