Abstract
This paper describes an approach to synthesis of efficient out-of-core code for a class of imperfectly nested loops that represent tensor contraction computations. Tensor contraction expressions arise in many accurate computational models of electronic structure. The developed approach combines loop fusion with loop tiling and uses a performance-model driven approach to loop tiling for the generation of out-of-core code. Experimental measurements are provided that show a good match with model-based predictions and demonstrate the effectiveness of the proposed algorithm.
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.
Similar content being viewed by others
References
Baumgartner, G., Bernholdt, D.E., Cociorva, D., Harrison, R., Hirata, S., Lam, C., Nooijen, M., Pitzer, R., Ramanujam, J., Sadayappan, P.: A High-Level Approach to Synthesis of High- Performance Codes for Quantum Chemistry. In: Proc Supercomputing 2002 (November 2002)
Cociorva, D., Baumgartner, G., Lam, C., Sadayappan, P., Ramanujam, J., Nooijen, M., Bernholdt, D., Harrison, R.: Space-Time Trade-Off Optimization for a Class of Electronic Structure Calculations. In: Proc. of ACM SIGPLAN 2002 Conference on Programming Language Design and Implementation (PLDI), June 2002, pp. 177–186 (2002)
Cociorva, D., Gao, X., Krishnan, S., Baumgartner, G., Lam, C., Sadayappan, P., Ramanujam, J.: Global Communication Optimization for Tensor Contraction Expressions under Memory Constraints. In: Proc. of 17th International Parallel & Distributed Processing Symposium (IPDPS) (April 2003)
Cociorva, D., Wilkins, J., Baumgartner, G., Sadayappan, P., Ramanujam, J., Nooijen, M., Bernholdt, D.E., Harrison, R.: Towards Automatic Synthesis of High-Performance Codes for Electronic Structure Calculations: Data Locality Optimization. In: Monien, B., Prasanna, V.K., Vajapeyam, S. (eds.) HiPC 2001. LNCS, vol. 2228, pp. 237–248. Springer, Heidelberg (2001)
Cociorva, D., Wilkins, J., Lam, C.-C., Baumgartner, G., Sadayappan, P., Ramanujam, J.: Loop optimization for a class of memory-constrained computations. In: Proc. 15th ACM International Conference on Supercomputing, Sorrento, Italy, June 2001, pp. 500–509 (2001)
Lam, C.: Performance Optimization of a Class of Loops Implementing Multi-Dimensional Integrals, Ph.D. Dissertation, The Ohio State University, Columbus, OH (August 1999)
Lam, C., Cociorva, D., Baumgartner, G., Sadayappan, P.: Optimization of Memory Usage and Communication Requirements for a Class of Loops Implementing Multi-Dimensional Integrals. In: Proc. 12th LCPC Workshop, San Diego, CA (August 1999)
Lam, C., Cociorva, D., Baumgartner, G., Sadayappan, P.: Memory-optimal evaluation of expression trees involving large objects. In: Proc. Intl. Conf. on High Perf. Comp. (December 1999)
Lam, C., Sadayappan, P., Wenger, R.: On Optimizing a Class of Multi-Dimensional Loops with Reductions for Parallel Execution. Par. Proc. Lett. 2(7), 157–168 (1997)
Lam, C., Sadayappan, P., Wenger, R.: Optimization of a Class of Multi-Dimensional Integrals on Parallel Machines. In: Proc. of Eighth SIAM Conf. on Parallel Processing for Scientific Computing, Minneapolis, MN (March 1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Krishnan, S. et al. (2003). Data Locality Optimization for Synthesis of Efficient Out-of-Core Algorithms. In: Pinkston, T.M., Prasanna, V.K. (eds) High Performance Computing - HiPC 2003. HiPC 2003. Lecture Notes in Computer Science, vol 2913. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-24596-4_44
Download citation
DOI: https://doi.org/10.1007/978-3-540-24596-4_44
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20626-2
Online ISBN: 978-3-540-24596-4
eBook Packages: Springer Book Archive