[Nek5000-users] Help With Basic NEK Configuration

[nek5000-users at lists.mcs.anl.gov]

Stefan & Paul,

Thank you both for all the hints. I will make use of these in my runs and  
will ask more questions if I have any.

Regards,
Harish



On Fri, 05 Aug 2011 09:08:01 -0500, <nek5000-users at lists.mcs.anl.gov>  
wrote:

>
> Harish,
>
> We can help you with the larger runs.
>
> The memory constraint is only one constraint, and it does
> not limit how _small_ a problem can be.
>
> There is a different constraint on problem size/processor
> count, which is that you must have at least one element
> per processor.   So if the problem has E=512 elements,
> then P_max = 512.
>
> It sounds like you've done a pretty good job of exploring
> the parameter space for the memory footprint etc.
>
> Memory requirements on a processor for a given compilation
> is, to leading order, given by the product:
>
>       S*lx1*ly1*lz1*lelt
>
> where S is between 200 and 400 words, each word being 8 bytes.
>
> There are other terms that can be important in some specific
> cases, particularly for large element (and hence, processor)
> counts.   But these are pretty well under control.   Specifically,
> there are two (4 byte) integer arrays that scale as lelg.
>
> Your total problem size (E==nelt in the .rea file) is constrained
> by:
>
>      E =< min(lelg,P*lelt)
>
> where P is the number of processors at run time.   Similarly, P
> is constrained by:
>
>      P =< lp
>
>
> There should be no memory impact for having a large value of lp.
>
> Paul
>
>
>
> On Thu, 4 Aug 2011, nek5000-users at lists.mcs.anl.gov wrote:
>
>> Hi,
>>
>> I am a new to NEK and I am trying to run a few examples for my OS  
>> scaling
>> experiments. I went through the manual which is at:
>> http://www.mcs.anl.gov/~fischer/new.pdf. In the chapter that speaks  
>> about
>> a basic problem setup, there is a line in the 'Memory Requirements'
>> section that says:
>>
>> Total Memory Required for simulation = E(N +1)^3 points × 400(wds/pt)  
>> ×
>> 8bytes/wd
>>
>> I have been playing around with the SIZE file a little bit and  
>> adjusting,
>> lx1, lelt, lxd, lp and lelg parameters. However I am unable to find  
>> where
>> I can adjust 'E' from the equation. I understand that 'lelt' is maximum
>> elements per processor and 'lelg' is maximum elements per simulation? Is
>> 'lelg' same as 'E'? is 'lelt = lelg / nprocs' ? I assumed that 'E =  
>> lelg'
>> and changed 'lelg' for 'timing' example to run it on a 4096 node
>> configuration (1 rack of BG/P:Intrepid in VN mode). I keep getting the
>> error:
>>
>> Too many processors ( 4096 ) for to few elements ( 512 )
>>
>> ... and the value 512 does not seem to change although I have changed
>> 'lelg' and run 'makenek timing'. Is the value of 'E' picked up from the
>> .re2 file? Would I have to generate it using 'genbox' after I change  
>> some
>> other configuration?
>>
>> I understand that NEK has been scaled upto 40 racks of Intrepid at
>> Argonne. I am also trying to run experiments of upto 32 racks if  
>> possible.
>> But I want to get started with smaller runs. Is there a repository of
>> configurations or tarred files where I could get larger, successfully  
>> run
>> problem instances?
>>
>> Thanks.
>>
>> Regards,
>> Harish Naik
>> Predoctoral Appointee
>> Mathematics & Computer Science Division
>> Argonne National Laboratory
>> P: (630) 252-2029
>> W: http://www.mcs.anl.gov/~hnaik
>> _______________________________________________
>> Nek5000-users mailing list
>> Nek5000-users at lists.mcs.anl.gov
>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users