A CUDA-MPI Hybrid Bitonic Sorting Algorithm For GPU Clusters
Document Type
Conference Proceeding
Publication Date
2012
Published In
41st International Conference On Parallel Processing Workshops
Abstract
We present a hybrid CUDA-MPI sorting algorithm that makes use of GPU clusters to sort large data sets. Our algorithm has two phases. In the first phase each node sorts a portion of the data on its GPU using a parallel bitonic sort. In the second phase the sorted subsequences are merged together in parallel using a reduction sorting network implemented in MPI across the cluster nodes. Performance results comparing our sorting algorithm to sequential quick sort yield speed-up values of up to 9.8 for sorting 4GB of data on a 32 node GPU cluster. We anticipate even better speed-up values using our algorithm on larger data sets and larger sized clusters.
Published By
IEEE
Conference
41st International Conference On Parallel Processing Workshops
Conference Dates
September 10-13, 2012
Conference Location
Pittsburgh, PA
Recommended Citation
Sam T. White , '14; Niels J. Verosky , '14; and Tia Newhall.
(2012).
"A CUDA-MPI Hybrid Bitonic Sorting Algorithm For GPU Clusters".
41st International Conference On Parallel Processing Workshops.
DOI: 10.1109/ICPPW.2012.82
https://works.swarthmore.edu/fac-comp-sci/115