Abstract
In this work we propose a run-time approach for the efficient parallel execution of doacross loops with indirect array accesses by means of a graph partitioning strategy. Our approach focuses not only on extracting parallelism among iterations of the loop, but also on exploiting data access locality to improve memory hierarchy behavior and thus the overall program speedup. The effectiveness of our algorithm is assessed in an SGI Origin 2000.
This work has been supported by the Ministry of Science and Technology of Spain and FEDER funds of the European Union (ref. TIC2001-3694-C02)
Chapter PDF
Similar content being viewed by others
References
D.-K. Chen, J. Torrellas and P.-C. Yew: An Efficient Algorithm for the Run-Time Parallelization of DOACROSS Loops, Proc. Supercomputing Conf. (1994) 518–527
J.H. Saltz, R. Mirchandaney and K. Crowley: Run-Time Parallelization and Scheduling of Loops, IEEE Trans. on Computers 40(5) (1991) 603–612
C.-Z. Xu and V. Chaudhary: Time Stamp Algorithms for Runtime Parallelization of DOACROSS Loops with Dynamic Dependences, IEEE Trans. on Parallel and Distributed Systems 12(5) (2001) 433–450
C.-Q. Zhu and P.-C. Yew: A Scheme to Enforce Data Dependence on Large Multiprocessor Systems, IEEE Trans. on Soft. Eng. 13(6) (1987) 726–739
C. Xu: Effects of Parallelism Degree on Run-Time Parallelization of Loops, Proc. 31st Hawaii Int. Conf. on System Sciences (1998)
G. Karypis and V. Kumar: A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs, SIAM J. on Scientific Comp. 20(1) (1999) 359–392
M.J. Martín, D.E. Singh, J. Touriño and F.F. Rivera: Exploiting Locality in the Run-time Parallelization of Irregular Loops, Proc. 2002 Int. Conf. on Parallel Processing (2002)
Y. Saad: SPARSKIT: a Basic Tool Kit for Sparse Matrix Computations (Version 2), at rs http://www.cs.umn.edu/Research/darpa/SPARSKIT/sparskit.html URL (1994)
I.S. Duff, R.G. Grimes and J.G. Lewis: User’s Guide for the Harwell-Boeing Sparse Matrix Collection, Tech. Report TR-PA-92-96, CERFACS (1992)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Martín, M.J., Singh, D.E., Touriño, J., Rivera, F.F. (2002). Improving Locality in the Parallelization of Doacross Loops. In: Monien, B., Feldmann, R. (eds) Euro-Par 2002 Parallel Processing. Euro-Par 2002. Lecture Notes in Computer Science, vol 2400. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45706-2_36
Download citation
DOI: https://doi.org/10.1007/3-540-45706-2_36
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44049-9
Online ISBN: 978-3-540-45706-0
eBook Packages: Springer Book Archive