[gmx-users] Re: PowerMac Quad Performance (Resolved)

[Tyler Luchko]

The -nice flag was, indeed, the problem.  Setting -nice 0 did not  
improve the already good results on my 2.7 GHz G5 but gave me the  
results I was looking for on the quad G5.  However, both mdrun and  
top claimed with the default -nice 19 that 100% cpu was being used on  
the quad.  There were no other cpu intensive processes to interfere  
with the benchmark.

Thanks for the help,

Tyler

> Tyler Luchko wrote:
>> Hello everybody,
>>
>> Has anyone had any experience running GROMACS on a PowerMac Quad
>> machine? We have purchased one as a test but have seen very poor
>> performance when running the benchmarks.  For example, I have   
>> compared
>> the results of the d.villin benchmark from the web with that  on  
>> my dual
>> 2.7 GHz PowerMac and our quad 2.5 GHz PowerMac:
>>
>> 2.5 GHz G5 (gromacs.org, v-3.2.1):            15309 ps/day
>> 2.7 GHz G5 (my dual 2.7,compiled w/ gcc-3.3, v-3.3):    15451 ps/day
>> 2.7 GHz G5 (my dual 2.7,precompiled, v-3.3):        13492 ps/day
>> 2.5 GHz G5 (my quad 2.5,compiled w/ gcc-3.3, v-3.3):     7516 ps/day
>> 2.5 GHz G5 (my quad 2.5,precompiled, v-3.3):         6475 ps/day
>> 2.5 GHz G5 x 4 (my quad 2.5,compiled w/ gcc-3.3, v-3.3):28800 ps/day
>>
>> As you can see the Quad is not doing well with the GROMACS code.  I
>> have tried the benchmark tool XBench and my own very simple C  
>> code  but
>> have not reproduced these problems.  Any insights or suggestion   
>> would
>> be greatly appreciated.
>>
> Check the log file to see whether it uses the Altivec code. Do also
> check that you have turned off the nice feature: mdrun -nice 0  
> otherwise
> your job will run at ow priority.
>
>
>
>> Thank you,
>>
>> Tyler
>>
>>
>>
>>  ________________________________________________________________
>> (_    Tyler Luchko                           Ph.D. Candidate    _)
>>  _)   Department of Physics            University of Alberta   (_
>> (_    Edmonton, Alberta, Canada                                 _)
>>  _)   780-492-1063                       tluchko at ualberta.ca   (_
>> (________________________________________________________________)
>>
>>
>>
>> _______________________________________________
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> Please don't post (un)subscribe requests to the list. Use the www
>> interface or send it to gmx-users-request at gromacs.org.
>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
>
> -- 
> David.
> ______________________________________________________________________ 
> __
> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,  	75124 Uppsala, Sweden
> phone:	46 18 471 4205		fax: 46 18 511 755
> spoel at xray.bmc.uu.se	spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 
> ++++
>