Scratch that. Apparently the script that was written to run this
benchmark (I didn't write it) was used under csh, which seems to have a
problem with redirecting a file as input to an MPI application. I
don't know exactly why, but when I rewrote the script in sh, it is
actually running.
--
Jonathan Herriott
Architecture and Performance Group
Apple Computer, Inc.
On Feb 7, 2005, at 1:29 PM, Jonathan Herriott wrote:
> This seems like the same problem I'm having when redirecting input
> from a file to standard input for the program to read. It doesn't
> seem to be recognizing it at all.
>
> --
> Jonathan Herriott
> Architecture and Performance Group
> Apple Computer, Inc.
>
> On Feb 7, 2005, at 5:38 AM, Yu Chen wrote:
>
>>>> Hi, Eiso
>>>> I suppose you are using Cyana, the format I believe should be:
>>>> cyana -t intel-lam -c 'mpirun -x CYANALIB -np 5' << EOF
>>>> ...
>>>> ...
>>>> EOF
>>> I know, but that has the same problem in my hands.
>>
>> Ok. I think it might be a problem with cyana too. so I cross-posted
>> this to Cyana list too.
>>
>> Yeah, we had this problem too, but not always, sometime it will
>> launch without errors, but just sit there doing nothing. We are
>> testing different settings.
>>
>> Has anybody from Cyana list successfully using Cyana-2.0, LAM-MPI
>> 7.x.x with Intel compilers?
>>
>> Chen
>>
>>
>>>
>>>> it's cyana's responsbility to take care of the input. Although I
>>>> don't know what version of Cyana you are using,we recently upgraded
>>>> both OS and Cyana, and seems Cyana stopped working as it suppose
>>>> to. We are using Cyana-2.0 on RHEL-AS-3 with Intel compiler and
>>>> LAM-MPI. We are still trying to figure it out.
>>> I'm using cyana-2.0, lam 7.1.1 ( compiled with intel ifort [8.0] and
>>> ssh
>>> on suse9.1
>>>
>>> Eiso
>>>
>>>
>>>> Chen
>>>> On Sun, 6 Feb 2005, Eiso AB wrote:
>>>>> I'm trying to use this to submit a parallel job via mpi:
>>>>> lamboot ....
>>>>> mpirun -x CYANALIB -np 5 cyana -t intel-lam << EOF
>>>>> CALC
>>>>> quit
>>>>> EOF
>>>>> lamhalt
>>>>> but the << EOF
>>>>> CALC
>>>>> quit
>>>>> EOF
>>>>> part doesn't seem to get passed through to cyana,
>>>>> because all I get is the 'cyana>' prompt and I can go
>>>>> on to type the input by hand.
>>>>> I've looked in the mpirun manpage and tried -f but that gives
>>>>> the same result.
>>>>> I'm using ssh for logging in without password.
>>>>> -Eiso
>>>
>>>
>>>
>>
>>
>> ===========================================
>> Yu Chen
>> Howard Hughes Medical Institute
>> Chemistry Building, Rm 182
>> University of Maryland at Baltimore County
>> 1000 Hilltop Circle
>> Baltimore, MD 21250
>>
>> phone: (410)455-6347 (primary)
>> (410)455-2718 (secondary)
>> fax: (410)455-1174
>> email: chen_at_[hidden]
>> ===========================================
>> _______________________________________________
>> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>>
>
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
|
