Article

Support for OpenMP tasks on cell architecture

Authors:

Shigang LiAuthors Info & Claims

ICA3PP'10: Proceedings of the 10th international conference on Algorithms and Architectures for Parallel Processing - Volume Part II

Pages 308 - 317

https://doi.org/10.1007/978-3-642-13136-3_32

Published: 21 May 2010 Publication History

Abstract

OpenMP task is the most significant feature in the new specification, which provides us with a way to handle unstructured parallelism. This paper presents a runtime library of task model on Cell heterogeneous multicore, which attempts to maximally utilize architectural advantages. Moreover, we propose two optimizations, an original scheduling strategy and an adaptive cut-off technique. The former combines breadth-first with the work-first scheduling strategy. While the latter adaptively chooses the optimal cut-off technique between max number of tasks and max task recursion level according to application characteristics. Performance evaluations indicate that our scheme achieves a speedup factor from 3.4 to 7.2 compared to serial executions.

References

[1]

Frigo, M., Leiserson, C.E., Randall, K.H.: The Implementation of the Cilk-5 Multithreaded Language. In: ACM SIGPLAN conference on Programming language design and implementation, pp. 212-223. ACM Press, New York (1998).

[2]

Reinders, J.: Intel Threading Building Blocks. Technical report, O'Reilly Media Inc. (2007).

[3]

T.X.D. Team: Report on the Experimental Language X10. Technical report, IBM (2006).

[4]

Chamberlain, B., Callahan, D., Zima, H.: Parallel programmability and the chapel language. J. Int. J. High Perform. Comput. Appl. 21, 291-312 (2007).

[5]

The Fortress Language Specification. Version 1.0 B (2007).

[6]

OpenMP Application Program Interface, Version 3.0. OpenMP Architecture Review Board (2008).

[7]

Duran, A., Corbalán, J., Ayguadé, E.: Evaluation of OpenMP task scheduling strategies. In: Eigenmann, R., de Supinski, B.R. (eds.) IWOMP 2008. LNCS, vol. 5004, pp. 101-110. Springer, Heidelberg (2008).

[8]

Shah, S., Haab, G., Petersen, P., Throop, J.: Flexible Control Structures for Parallelism in OpenMP. In: 1st European Workshop OpenMP, pp. 1219-1239 (1999).

[9]

Teruel, X., Martorell, X., Duran, A., Ferrer, R., Ayguadé, E.: Support for OpenMP Tasks in Nanos v4. In: Proc. Conf. Center for Advanced Studies on Collaborative Research, pp. 256-259. ACM Press, New York (2007).

[10]

Teruel, X., Unnikrishnan, P., Martorell, X., et al.: Openmp tasks in ibm XL compilers. In: Proc. of the 2008 conference of the center for advanced studies on collaborative research, pp. 207-221. ACM Press, New York (2008).

[11]

Altevogt, P.: IBM BladeCenter QS21 Hardware Performance. IBM Technical White Paper WP101245 (2008).

[12]

Leijen, D., Hall, J.: Optimize Managed Code for Multi-Core Machines. J. MSDN Magazine, 1098-1116 (2007).

[13]

Leijen, D., Schulte, W., Burckhardt, S.: The design of a task parallel library. In: International Conference on Object Oriented Programming, Systems, Languages and Applications, pp. 227-242. ACM Press, New York (2009).

[14]

Balart, J., Duran, A., Gonza'lez, M., Martorell, X., et al.: Nanos Mercurium: A Research Compiler for OpenMP. In: 6th European Workshop OpenMP, pp. 103-109 (2004).

[15]

Ayguadé, E., Duran, A., Hoeflinger, J., et al.: An Experimental Evaluation of the New OpenMP Tasking Model. In: Adve, V., Garzarán, M.J., Petersen, P. (eds.) LCPC 2007. LNCS, vol. 5234, pp. 63-77. Springer, Heidelberg (2008).

[16]

Cody, A., James, L., Lei, H., Barbara, C.: OpenMP 3.0 Tasking Implementation in OpenUH. In: 2nd Open64 Workshop at CGO (2009).

[17]

Rico, A., Ramirez, A., Valero, M.: Available task-level parallelism on the cell BE. J. Scientific Programming 17, 59-76 (2009).

[18]

Bellens, P., Perez, J.M., Badia, R.M., Labarta, J.: CellSs: a programming model for the Cell BE Architecture. In: Proc. of the 2006 ACM/IEEE Conference on Supercomputing. ACM Press, New York (2006).

[19]

Certner, O., Li, Z., Palatin, P., et al.: A Practical Approach for Reconciling High and Predictable Performance in Non-Regular Programs. In: 1st Workshop on Programmability Issues for Multi-Core Computers, pp. 740-745. ACM Press, New York (2008).

[20]

Duran, A., Corbalán, J., Ayguadé, E.: An adaptive cut-off for task parallelism. In: Proc. of the 2008 ACM/IEEE Conf. on Supercomputing, pp. 1-11. IEEE Press, Los Alamitos (2008).

[21]

Martorell, X., Labarta, J., Navarro, N., Ayguadé, E.: Nano-Threads Library Design, Implementation and Evaluation. Technical Report UPC-DAC-1995-33, DAC/UPC (1995).

[22]

Cong, G., Kodali, S., Krishnamoorthy, S., et al.: Solving large, irregular graph problems using adaptive work-stealing. In: Proc. of the International Conference on Parallel Processing, pp. 536-545. IEEE Press, New York (2008).

Cited By

Wang CChandrasekaran SSun PChapman BHolt J(2013)Portable mapping of openMP to multicore embedded systems using MCA APIsACM SIGPLAN Notices10.1145/2499369.246556948:5(153-162)Online publication date: 20-Jun-2013
https://dl.acm.org/doi/10.1145/2499369.2465569
Wang CChandrasekaran SSun PChapman BHolt JFranke BXue J(2013)Portable mapping of openMP to multicore embedded systems using MCA APIsProceedings of the 14th ACM SIGPLAN/SIGBED conference on Languages, compilers and tools for embedded systems10.1145/2491899.2465569(153-162)Online publication date: 20-Jun-2013
https://dl.acm.org/doi/10.1145/2491899.2465569
Wang CChandrasekaran SSun PChapman BHolt JFranke BXue J(2013)Portable mapping of openMP to multicore embedded systems using MCA APIsProceedings of the 14th ACM SIGPLAN/SIGBED conference on Languages, compilers and tools for embedded systems10.1145/2465554.2465569(153-162)Online publication date: 20-Jun-2013
https://dl.acm.org/doi/10.1145/2465554.2465569
Show More Cited By

Support for OpenMP tasks on cell architecture
1. General and reference
  1. Cross-computing tools and techniques

Recommendations

Barcelona OpenMP Tasks Suite: A Set of Benchmarks Targeting the Exploitation of Task Parallelism in OpenMP
ICPP '09: Proceedings of the 2009 International Conference on Parallel Processing

Traditional parallel applications have exploited regular parallelism, based on parallel loops. Only a few applications exploit sections parallelism. With the release of the new OpenMP specification (3.0), this programming model supports tasking. ...
Investigating Dependency Graph Discovery Impact on Task-based MPI+OpenMP Applications Performances
ICPP '23: Proceedings of the 52nd International Conference on Parallel Processing

The architecture of supercomputers is evolving to expose massive parallelism. MPI and OpenMP are widely used in application codes on the largest supercomputers in the world. The community primarily focused on composing MPI with OpenMP before its version ...
Extending synchronization constructs in openMP to exploit pipeline parallelism on heterogeneous multi-core
ICA3PP'11: Proceedings of the 11th international conference on Algorithms and architectures for parallel processing - Volume Part II

The ability of expressing multiple-levels of parallelism is one of the significant features in OpenMP parallel programming model. However, pipeline parallelism is not well supported in OpenMP. This paper proposes extensions to OpenMP directives, aiming ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image Guide Proceedings

ICA3PP'10: Proceedings of the 10th international conference on Algorithms and Architectures for Parallel Processing - Volume Part II

May 2010

467 pages

ISBN:3642131352

Editors:
Ching-Hsien Hsu
Department of Computer Science and Information Engineering, Chung Hua University, Hsinchu, Taiwan, China
,
Laurence T. Yang
Department of Computer Science, St. Francis Xavier University, Antigonish, NS, Canada
,
Jong Hyuk Park
Department of Computer Science ad Engineering, Seoul National University of Technology, 172 Gongreund 2-dong, Nowon-gou, Seoul, NS, Korea
,
Sang-Soo Yeo
Division of Computer Engineering, Mokwon University, 172 Gongreund 2-dong, Nowon-gou, Daejeon, NS, Korea

Publisher

Springer-Verlag

Berlin, Heidelberg

Publication History

Published: 21 May 2010

Author Tags

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

5
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 27 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

Wang CChandrasekaran SSun PChapman BHolt J(2013)Portable mapping of openMP to multicore embedded systems using MCA APIsACM SIGPLAN Notices10.1145/2499369.246556948:5(153-162)Online publication date: 20-Jun-2013
https://dl.acm.org/doi/10.1145/2499369.2465569
Wang CChandrasekaran SSun PChapman BHolt JFranke BXue J(2013)Portable mapping of openMP to multicore embedded systems using MCA APIsProceedings of the 14th ACM SIGPLAN/SIGBED conference on Languages, compilers and tools for embedded systems10.1145/2491899.2465569(153-162)Online publication date: 20-Jun-2013
https://dl.acm.org/doi/10.1145/2491899.2465569
Wang CChandrasekaran SSun PChapman BHolt JFranke BXue J(2013)Portable mapping of openMP to multicore embedded systems using MCA APIsProceedings of the 14th ACM SIGPLAN/SIGBED conference on Languages, compilers and tools for embedded systems10.1145/2465554.2465569(153-162)Online publication date: 20-Jun-2013
https://dl.acm.org/doi/10.1145/2465554.2465569
Wang CChandrasekaran SChapman BHolt J(2013)libEOMPProceedings of the 2013 International Workshop on Programming Models and Applications for Multicores and Manycores10.1145/2442992.2443001(83-92)Online publication date: 23-Feb-2013
https://dl.acm.org/doi/10.1145/2442992.2443001
Li SYao SHe HSun LChen YPeng Y(2011)Extending synchronization constructs in openMP to exploit pipeline parallelism on heterogeneous multi-coreProceedings of the 11th international conference on Algorithms and architectures for parallel processing - Volume Part II10.5555/2075462.2075469(54-63)Online publication date: 24-Oct-2011
https://dl.acm.org/doi/10.5555/2075462.2075469

View Options

View options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

Media

Figures

Other

Tables

View Table of Contents