- From: Terje Bless <link@pobox.com>
- Date: Wed, 30 Jul 2003 14:48:01 +0200
- To: Ville Skyttä <ville.skytta@iki.fi>
- cc: QA Dev <public-qa-dev@w3.org>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Ville Skyttä <ville.skytta@iki.fi> wrote:
>Yep, but adding roughly 0.3-0.4 seconds CPU time to each request can't
>be good... dunno what the overall effect would be in per-request
>wallclock time in production at v.w.o.
link@validator:cgi-bin<1>$ time HEAD \
http://validator.w3.org/check?uri=http://www.w3.org/ >/dev/null
real 0m1.159s
user 0m0.340s
sys 0m0.010s
link@validator:cgi-bin<2>$ time HEAD \
http://validator.w3.org/check?uri=http://www.yahoo.com/ >/dev/null
real 0m1.739s
user 0m0.280s
sys 0m0.060s
So we're talking about between 1.0 and 1.75 wallclock seconds for the
processing alone; making your numbers something like 30-50% increase.
>Will there be another live beta release? Could estimate the cost there.
I'm aiming to do a new beta as soon as I make a final call on some WS code I'm
working on (haven't quite decided on Go/No Go yet) && Olivier is happy with
docs and style tweaks && I hear back from Jim about some JavaScript stuff[0].
Lets make a point to profile this stuff in a more reliable manner then. Well,
or you could just go ahead immediately in a separate server on v.w3.org.
[0] - I've "voluntered" Jim Ley to do some JavaScript for the Results page to
show Error Explanations on demand; but he's pretty busy so I don't know
how long until he has time for this. If it's soon I want to hold off on
beta until we can add the JS to maximized the amount of testing it gets.
- --
Interviewer: "In what language do you write your algorithms?"
Abigail: English.
Interviewer: "What would you do if, say, Telnet didn't work?"
Abigail: Look at the error message.
-----BEGIN PGP SIGNATURE-----
Version: PGP SDK 3.0.2
iQA/AwUBPye+f6PyPrIkdfXsEQLorgCglVLO7B2IGC0qtj1SZk4BuW07IXAAoJpo
W6rFLBTXDgesvg4+v5EujdeC
=Zb7k
-----END PGP SIGNATURE-----
Received on Wednesday, 30 July 2003 08:48:13 UTC