Page 2 of 2
Re: How to fully utilize systems?
Posted: Sun Aug 22, 2010 5:31 pm
by Nathan_P
zoharek wrote:OK, tomorrow I will be setting new machine for bigadv, wish me luck, it's 4x Xeon X7560 with HT plus 64GB 1066MHz RAM, hope it will be enough, because it's most I can get from single machine, next would be clusters only.
Also I wonder about GPU clients, as I previously stated, I run 2x Tesla C1060, each makes about 8000 PPD, however I have read that GTX460 goes as far as 10000 PPD. I could replace Teslas with 2x GTX460 (as long as they need 2x 6pin power connectors). This is weird to me though, I thought Teslas > GTX.
This is going to be insane - in a good way
Re: How to fully utilize systems?
Posted: Mon Aug 23, 2010 1:09 pm
by zoharek
Hah, small update, my friend did start another smp client on dual xeon server, that is why I have had such big frametimes. I run it now again with -bigadv and I have
39:40min TPF, with estimated PPD of
24520 
Quad Xeon is already built, but need running some stress tests before deploying
Another update, we screwed up configuration a little bit, and need to correct it, so Quad Xeon still inoperative, next update tomorrow.
25.08. Update: Motherboard for Quad Xeon gone to hell, awaiting new one.
Re: How to fully utilize systems?
Posted: Sat Aug 28, 2010 6:02 pm
by zoharek
Bumping the thread for people interested. Quad Xeon is finally up and running, I just set it up, it started to process project 2685 (R4, C18, G16), after first percents being processed no yay yet, maybe it will speed up later:
PPD 85664
TPF 17m17s
Credit 102817
To remind:
4x Xeon X7560
64GB 1066MHz RAM
Machine is dedicated for F@H
I will check configuration, and see what can I do to speed it up. Note, it's running Win Server 2008 R2 Database.
Re: How to fully utilize systems?
Posted: Mon Aug 30, 2010 12:35 pm
by zoharek
OK, it was running on real cores only with -smp switch. I forced all cores with -smp 64 and got to 14m13s per percent on 2685 project (one of slower ones). Good