Adapting Irregular Computations to Large CPU-GPU Clusters in the MADNESS Framework

Vlad Slavici, Raghu Varier, Gene Cooperman, Robert J. Harrison
Northeastern University, Boston, MA
IEEE International Conference on Cluster Computing (CLUSTER), 2012


   title={Adapting Irregular Computations to Large CPU-GPU Clusters in the MADNESS Framework},

   author={Slavici, V. and Varier, R. and Cooperman, G. and Harrison, R.J.},

   booktitle={Cluster Computing (CLUSTER), 2012 IEEE International Conference on},





Download Download (PDF)   View View   Source Source   



Graphics Processing Units (GPUs) are becoming the workhorse of scalable computations. MADNESS is a scientific framework used especially for computational chemistry. Most MADNESS applications use operators that involve many small tensor computations, resulting in a less regular organization of computations on GPUs. A single GPU kernel may have to multiply by hundreds of small square matrices (with fixed dimension ranging from 10 to 28). We demonstrate a scalable CPU-GPU implementation of the MADNESS framework over a 500-node partition on the Titan supercomputer. For this hybrid CPU-GPU implementation, we observe up to a 2.3-times speedup compared to an equivalent CPU-only implementation with 16 cores per node. For smaller matrices, we demonstrate a speedup of 2.2-times by using a custom CUDA kernel rather than a cuBLAS-based kernel.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: