Abstract
We study the impact of asynchronism on parallel iterative algorithms in the particular context of local clusters of workstations including GPUs. The application test is a classical PDE problem of advection-diffusion-reaction in 3D. We propose an asynchronous version of a previously developed PDE solver using GPUs for the inner computations. The algorithm is tested with two kinds of clusters, a homogeneous one and a heterogeneous one (with different CPUs and GPUs).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Amitai, D., Averbuch, A., Israeli, M., Itzikowitz, S.: Implicit-explicit parallel asynchronous solver for PDEs. SIAM J. Sci. Comput. 19, 1366–1404 (1998)
Bahi, J., Contassot-Vivier, S., Couturier, R.: Evaluation of the asynchronous iterative algorithms in the context of distant heterogeneous clusters. Parallel Computing 31(5), 439–461 (2005)
Bahi, J., Contassot-Vivier, S., Couturier, R.: An Efficient and Robust Decentralized Algorithm for Detecting the Global Convergence in Asynchronous Iterative Algorithms. In: Palma, J.M.L.M., Amestoy, P.R., Daydé, M., Mattoso, M., Lopes, J.C. (eds.) VECPAR 2008. LNCS, vol. 5336, pp. 251–264. Springer, Heidelberg (2008)
Baudet, G.M.: Asynchronous iterative methods for multiprocessors. J. ACM 25, 226–244 (1978)
Bertsekas, D.P., Tsitsiklis, J.N.: Parallel and Distributed Computation: Numerical Methods. Prentice Hall, Englewood Cliffs (1989)
Basic linear algebra subprograms, http://www.netlib.org/blas/
Bojanczyk, A.: Optimal asynchronous newton method for the solution of nonlinear equations. J. ACM 31, 792–803 (1984)
Bru, R., Migallon, V., Penadés, J., Szyld, D.B.: Parallel synchronous and asynchronous two-stage multisplitting methods. ETNA 3, 24–38 (1995)
Chazan, D., Miranker, W.: Chaotic relaxation. Linear Algebra Appl. 2, 199–222 (1969)
Cosnard, M., Fraignaud, P.: Analysis of asynchronous polynomial root finding methods on a distributed memory multicomputer. IEEE Trans. on Parallel and Distributed Systems 5(6) (June 1994)
Fletcher, R.: Conjugate gradient methods for indefinite systems. In: Watson, G. (ed.) Numerical Analysis. LNM, vol. 506, pp. 73–89. Springer, Heidelberg (1976), doi:10.1007/BFb0080116
Frommer, A., Mayer, G.: On the theory and practice of multisplitting mehods in parallel computation. Computing 49, 63–74 (1992)
Frommer, A., Szyld, D.B.: Asynchronous iterations with flexible communication for linear systems. Calculateurs Parallèles, Réseaux et Systèmes Répartis 10, 421–429 (1998)
Frommer, A., Szyld, D.B.: On asynchronous iterations. J. Comput. and Appl. Math. 123, 201–216 (2000)
Gonzalez, R., Horowitz, M.: Energy dissipation in general pupose microprocessors. IEEE Journal of Solid-State Circuits 31(9) (September 1996)
Heroux, M.A.: A proposal for a sparse blas toolkit. SPARKER working note #2. Cray research, Inc. (1992)
Hundsdorfer, W., Verwer, J.G.: Numerical Solution of Time-Dependent Advection-Diffusion-Reaction Equations, 1st edn. Springer Series in Computational Mathematics, vol. 33. Springer, Heidelberg (2003)
Jost, T., Contassot-Vivier, S., Vialle, S.: An efficient multi-algorithm sparse linear solver for GPUs. In: Parallel Computing: From Multicores and GPU’s to Petascale. Advances in Parallel Computing, vol. 19, pp. 546–553. IOS Press (2010)
Jost, T., Contassot-Vivier, S., Vialle, S.: On the interest of clusters of GPUs. In: Grid’5000 Spring School 2010, Lille, France (April 2010)
Miellou, J.-C.: Algorithmes de relaxation chaotique à retards. R.A.I.R.O. R 1, 55–82 (1975)
Szyld, D.B., Xu, J.: Convergence of partially asynchronous block quasi-newton methods for nonlinear systems of equations. J. Comp. and Appl. Math. 103, 307–321 (1999)
van der Vorst, H.A.: Bi-cgstab: A fast and smoothly converging variant of bi-cg for the solution of nonsymmetric linear systems. SIAM Journal on Scientific and Statistical Computing 13(2), 631–644 (1992)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Contassot-Vivier, S., Jost, T., Vialle, S. (2012). Impact of Asynchronism on GPU Accelerated Parallel Iterative Computations. In: Jónasson, K. (eds) Applied Parallel and Scientific Computing. PARA 2010. Lecture Notes in Computer Science, vol 7133. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-28151-8_5
Download citation
DOI: https://doi.org/10.1007/978-3-642-28151-8_5
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-28150-1
Online ISBN: 978-3-642-28151-8
eBook Packages: Computer ScienceComputer Science (R0)