QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: anjana on September 11, 2012, 04:53

Title: very slow calculation
Post by: anjana on September 11, 2012, 04:53

I am using the demo version of atk 12.2.2. I am running the atomistix in a cluster. But the calculation are really slow, as if in a normal computer. I tested the parallelization with the test_script file, and it is perfectly fine. What might be the problem? Is it because I am using the demo version in a cluster?? or something else???


Anjana

Title: Re: very slow calculation
Post by: Anders Blom on September 11, 2012, 19:23

A common error is over-parallelization by running one MPI process per core. ATK will only exhibit a positive performance boost from MPI if you limit the number of MPI processes to the number of available machines (or number of sockets, in case each machine has more than one). In this scenario ATK will use OpenMP to thread (parallelize) over the cores, although this is more limited compared to MPI in terms of performance improvement. But if you have too many MPI processes on a machine, they will fight for access to memory and the machine becomes overloaded with communication between the MPI processes. So, again, the simple rule for best performance is: one MPI process per socket.