Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130713011518.GA18741@openwall.com>
Date: Sat, 13 Jul 2013 05:15:18 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: new dev box wishes

On Tue, Jul 09, 2013 at 08:22:01PM +0400, Solar Designer wrote:
> On Fri, Jun 28, 2013 at 10:54:46PM +0400, Solar Designer wrote:
> > Our Xeon Phi 5110P should arrive on Tuesday
> 
> That was a week ago, and it did arrive.  Here are some pics of it,
> including with the blue shroud removed:
> 
> http://openwall.info/wiki/internal/xeon_phi
> 
> It didn't fit into a couple of server platforms that were readily
> available.  The next thing to try is likely installing some cooling fans
> intended for 1U chassis right onto the Xeon Phi card, thus producing
> airflow similar to what the card would be exposed to in proper chassis.
> Maybe one or two of these:
> 
> http://www.srv-trade.ru/catalog/3918429255/FAN-0086L/
> http://www.srv-trade.ru/catalog/3918429255/FAN-0086L4/
> 
> (12k RPM, so must be very noisy, but that's OK for this machine).

This was attempted:

http://openwall.info/wiki/internal/xeon_phi#An-attempt-at-cooling-it

but it turned out not to matter yet, because Xeon Phi simply doesn't
work in this motherboard:

http://forum.giga-byte.co.uk/index.php/topic,12801.0.html

For now, our new dev box has been setup with GPUs, but without Xeon Phi.
I am running some Bitcoin mining as a stress-test on its 3 GPUs at the
moment.  7990's first GPU got to 93C at first, but then I forced 7990's
fans speed to 100% (somehow it didn't go above 74% on its own) and the
temperatures dropped to under 90C.

solar@...l:~/j/bleeding-jumbo/run$ ./john -te -form=mscash2-opencl -dev=0,1,4
Device 0: Tahiti (AMD Radeon HD 7900 Series )
Optimal Work Group Size:128
Kernel Execution Speed (Higher is better):1.692624
Optimal Global Work Size:278528
Device 1: Tahiti (AMD Radeon HD 7900 Series )
Optimal Work Group Size:256
Kernel Execution Speed (Higher is better):1.692732
Optimal Global Work Size:557056
Device 4: GeForce GTX TITAN 
Optimal Work Group Size:1024
Kernel Execution Speed (Higher is better):1.040073
Optimal Global Work Size:729088
Benchmarking: mscash2-opencl, M$ Cache Hash 2 (DCC2) [PBKDF2-SHA1 OpenCL]... DONE
Raw:    273066 c/s real, 113212 c/s virtual

Alexander

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.