(Please don’t drop the discussion thread, added it back.) Is 1.0 second for one step or for the whole application time? Also, can you provide the following information of your program? 1. What’s the version of your MPI? 2. How many processes you use? 3. Can you give us a minimum working example of your code so that we can reproduce your problem? You can read a tutorial of parallel computing here: https://computing.llnl.gov/tutorials/parallel_comp/ — Huiwei On Apr 14, 2014, at 10:54 AM, harsh soni <[email protected]> wrote:

Dear Huiwei,

Thanks for the replay. I tried to calculate the time taken by the various segments in my code. I realize, the slowest part correspond to hug communication between cores, with communication of 15750 double precision pre core, at every step. The time taken by this communication is around 1.0 second. I just want to know, how the communication affects the speed of code. I read about "parallel slowdown". Can please tell me something about it.

Sorry for bothering you!

Thanks Harsh Soni Ph.D. student, IISc, Bengaluru (India)

On Thu, Apr 10, 2014 at 6:47 PM, Lu, Huiwei <[email protected]> wrote: How slow is that? What’s the version of your MPI? How many processes you use?

If only 0.3% RAM was used for each core, it was unlikely the code went slow because of running out of RAM. Can you give us a minimum working example of your code so that we can reproduce your problem?

— Huiwei Lu Postdoc Appointee Mathematics and Computer Science Division, Argonne National Laboratory http://www.mcs.anl.gov/~huiweilu/

On Apr 10, 2014, at 7:57 AM, harsh soni <[email protected]> wrote:

...

Hi, I am running a mpi code on cluster. It goes slow after some time. I don't allocate any array in main for loop, every array is allocated initially. Each core is showing 0.3 to 0.5 % RAM usage. doe sit go slow due to the RAM. I use fortran language.

Thanks

With best regards Harsh Soni Ph.D. student, IISc, Bengaluru (India)

_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss