search

ComputationalRadiationPhysics / haseongpu

HASEonGPU: High performance Amplified Spontaneous Emission on GPU
http://www.hzdr.de/crp
Other
7 stars 6 forks source link

Speed up sequential CPU code through OpenMP #51

Open slizzered opened 9 years ago

slizzered commented 9 years ago

There might be some loops that can pose a bottleneck. They might be rather easily parallelized with OpenMP (cmake-file will need to be tweaked)

slizzered commented 9 years ago

I started working on this. branch: https://github.com/slizzered/haseongpu/tree/issue51-openMP-hostcode

Not sure how necessary this actually is... might introduce code complexity without benefits. See also the commit message: ( https://github.com/slizzered/haseongpu/commit/91625acadf2c623667acb5b404a6138d12a7ff65 )

erikzenker commented 9 years ago

To gain maximal performance we need to reduce the runtime of our sequential code base. We have two possibilities to gain this reduction:

So, I think some investigation makes sense. But, you are right, it looks a bit weird if computation unrelated code is parallized with OpenMP :hibiscus:

slizzered commented 9 years ago

Well, it looks not too weird to me. The problem is rather, that it does not bring any speedup, since the loops are pretty small/fast.

The MATLAB functions seem to be one of the more important problems (really slow...)