Developing a High Performance Software Library with MPI and CUDA for Matrix Computations
Bogdan Oancea () and
Tudorel Andrei ()
Computational Methods in Social Sciences (CMSS), 2013, vol. 1, issue 2, 5-10
Nowadays, the paradigm of parallel computing is changing. CUDA is now a popular programming model for general purpose computations on GPUs and a great number of applications were ported to CUDA obtaining speedups of orders of magnitude comparing to optimized CPU implementations. Hybrid approaches that combine the message passing model with the shared memory model for parallel computing are a solution for very large applications. We considered a heterogeneous cluster that combines the CPU and GPU computations using MPI and CUDA for developing a high performance linear algebra library. Our library deals with large linear systems solvers because they are a common problem in the fields of science and engineering. Direct methods for computing the solution of such systems can be very expensive due to high memory requirements and computational cost. An efficient alternative are iterative methods which computes only an approximation of the solution. In this paper we present an implementation of a library that uses a hybrid model of computation using MPI and CUDA implementing both direct and iterative linear systems solvers. Our library implements LU and Cholesky factorization based solvers and some of the non-stationary iterative methods using the MPI/CUDA combination. We compared the performance of our MPI/CUDA implementation with classic programs written to be run on a single CPU.
Keywords: parallel algorithms; linear algebra; CUDA; MPI; GPU computing (search for similar items in EconPapers)
References: View references in EconPapers View complete reference list from CitEc
Citations Track citations by RSS feed
Downloads: (external link)
http://cmss.univnt.ro/wp-content/uploads/vol/split ... _issue_2_art_001.pdf First version, 2013 (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:ntu:ntcmss:vol1-iss2-13-05
Access Statistics for this article
Computational Methods in Social Sciences (CMSS) is currently edited by Bogdan Oancea
More articles in Computational Methods in Social Sciences (CMSS) from "Nicolae Titulescu" University of Bucharest, Faculty of Economic Sciences Contact information at EDIRC.
Series data maintained by Stefan Ciucu ().