From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Fri Jun 10 2011 - 06:25:45 CDT
Hi,
the real speedup from a number of cpus alone to the same number of cpus + 1GPU is the highest at 1:1. Most konfiurations bring better resuls for 2:2 as for 2:1 because there are already waiting times for the processes to access the gpu. Using multible cores per gpu is just a strategy to beeter utilize the gpu when its idle because the process needs to calc something on the cpu. Then another process can use the gpu in this time. You will speedup when using 6:1 over 1:1, but that’s only due to the further cpu power, not the gpu, so your machine with 48 cores is fast enough without a gpu ;)
Best regards
Norman Geist.
-----Ursprüngliche Nachricht-----
Von: Francesco Pietra [mailto:chiendarret_at_gmail.com]
Gesendet: Freitag, 10. Juni 2011 11:35
An: Norman Geist
Betreff: Re: namd-l: max number of CPU PEs per GPU?
On Fri, Jun 10, 2011 at 9:56 AM, Norman Geist
<norman.geist_at_uni-greifswald.de> wrote:
> Hi Danny,
>
>
>
> The speedup already falls from 1:1 to 2:1 CPU:GPU ratio.
What do you mean? That 1CPU per 1GPU is the best choice? Or am i
misunderstanding?
Thank you
chiendarret
>
>
> Norman Geist.
>
>
>
> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag
> von Danny Xu
> Gesendet: Freitag, 10. Juni 2011 08:18
> An: namd-l
> Betreff: namd-l: max number of CPU PEs per GPU?
>
>
>
> Hi,
>
>
>
> I was testing the namd 2.8b2 cuda binary (Linux-x86_64-CUDA) on a 48-core
> opteron/1x GTX480 server. I got:
>
> ....
>
> Pe 41 sharing CUDA device 0 first 0 next 42
>
> Pe 40 sharing CUDA device 0 first 0 next 41
>
> FATAL ERROR: CUDA error cudaStreamCreate on Pe 23 (chili device 0): no
> CUDA-capable device is available
>
> ------------- Processor 23 Exiting: Called CmiAbort ------------
>
> Reason: FATAL ERROR: CUDA error cudaStreamCreate on Pe 23 (chili device 0):
> no CUDA-capable device is available
>
>
>
> First I thought there was something wrong with the GTX480 card. Then I tried
> 12 PE's and the job ran just fine. It turned out the max number core I could
> use with one GTX480 is 14. My question is, is this hard coded in namd 2.8?
> Or it's a bug in 2.8b2 and has been fixed in 2.8?
>
>
>
> It'd be nice if all 48-core could be used with the GPU.
>
>
>
> Thanks,
>
>
>
> DX
This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:20:25 CST