Abstract
A loop with irregular assignment computations contains loop-carried output data dependences that can only be detected at run-time. In this paper, a load-balanced method based on the inspector-executor model is proposed to parallelize this loop pattern. The basic idea lies in splitting the iteration space of the sequential loop into sets of conflict-free iterations that can be executed concurrently on different processors. As will be demonstrated, this method outperforms existing techniques. Irregular access patterns with different load-balancing and reusability properties are considered in the experiments.
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
Arenaz, M., Touriño, J., Doallo, R.: Irregular assignment computations on cc-NUMA multiprocessors. In: Zima, H.P., Joe, K., Sato, M., Seo, Y., Shimasaki, M. (eds.) ISHPC 2002. LNCS, vol. 2327, pp. 361–369. Springer, Heidelberg (2002)
Arenaz, M., Touriño, J., Doallo, R.: A GSA-Based Compiler Infrastructure to Extract Parallelism from Complex Loops. In: Proceedings of 17th ACM International Conference on Supercomputing, ICS 2003, San Francisco, CA, pp. 193–204 (2003)
Glassner, A.: Graphics Gems. Academic Press, London (1993)
Gutiérrez, E., Plata, O., Zapata, E.L.: Balanced, Locality-Based Parallel Irregular Reductions. In: Proceedings of 14th International Workshop on Languages and Compilers for Parallel Computing, LCPC 2001, Cumberland Falls, KY (2001)
Han, H., Tseng, C.-W.: Efficient Compiler and Run-Time Support for Parallel Irregular Reductions. Parallel Computing 26(13-14), 1861–1887 (2000)
Knobe, K., Sarkar, V.: Array SSA Form and Its Use in Parallelization. In: Proceedings ACM SIGACT-SIGPLAN Symposium on the Principles of Programming Languages, pp. 107–120 (1998)
Lin, Y., Padua, D.A.: On the automatic parallelization of sparse and irregular fortran programs. In: O’Hallaron, D.R. (ed.) LCR 1998. LNCS, vol. 1511, pp. 41–56. Springer, Heidelberg (1998)
Martín, M.J., Singh, D.E., Touriño, J., Rivera, F.F.: Exploiting Locality in the Run-time Parallelization of Irregular Loops. In: Proceedings of 31st International Conference on Parallel Processing, ICPP 2002, Vancouver, Canada, pp. 27–34 (2002)
OpenMP Architecture Review Board: OpenMP: A Proposed Industry Standard API for Shared Memory Programming (1997)
Rauchwerger, L., Padua, D.A.: The LRPD Test: Speculative Run-Time Parallelization of Loops with Privatization and Reduction Parallelization. IEEE Transactions on Parallel and Distributed Systems 10(2), 160–180 (1999)
Saad, Y.: SPARSKIT: A Basic Tool Kit for Sparse Matrix Computations (1994), http://www.cs.umn.edu/Research/darpa/SPARSKIT/sparskit.html
Turek, S., Becker, C.: Featflow: Finite Element Software for the Incompressible Navier-Stokes Equations. User Manual (1998), http://www.featflow.de
Wolfe, M.J.: Optimizing Supercompilers for Supercomputers. Pitman, London and The MIT Press, Cambridge (1989)
Xu, C.-Z., Chaudhary, V.: Time Stamp Algorithms for Runtime Parallelization of DOACROSS Loops with Dynamic Dependences. IEEE Transactions on Parallel and Distributed Systems 12(5), 433–450 (2001)
Yu, H., Rauchwerger, L.: Adaptive Reduction Parallelization Techniques. In: Proceedings of the 14th ACM International Conference on Supercomputing, Santa Fe, NM, pp. 66–77 (2000)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Arenaz, M., Touriño, J., Doallo, R. (2004). An Inspector-Executor Algorithm for Irregular Assignment Parallelization. In: Cao, J., Yang, L.T., Guo, M., Lau, F. (eds) Parallel and Distributed Processing and Applications. ISPA 2004. Lecture Notes in Computer Science, vol 3358. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30566-8_4
Download citation
DOI: https://doi.org/10.1007/978-3-540-30566-8_4
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-24128-7
Online ISBN: 978-3-540-30566-8
eBook Packages: Computer ScienceComputer Science (R0)