Leveraging non-blocking collective communication in high-performance applications

Subscribe (Full Service)


	Search: The ACM Digital Library The Guide

	Feedback

Leveraging non-blocking collective communication in high-performance applications

Full text

Pdf (93 KB)

Source

ACM Symposium on Parallel Algorithms and Architectures archive
Proceedings of the twentieth annual symposium on Parallelism in algorithms and architectures table of contents

Munich, Germany

SESSION: Brief announcements table of contents

Pages 113-115

Year of Publication: 2008

ISBN:978-1-59593-973-9

Authors

Torsten Hoefler	Indiana University, Bloomington, IN, USA
Peter Gottschling	Indiana University, Bloomington, IN, USA
Andrew Lumsdaine	Indiana University, Bloomington, IN, USA

Sponsors

ACM: Association for Computing Machinery
SIGACT: ACM Special Interest Group on Algorithms and Computation Theory
SIGARCH: ACM Special Interest Group on Computer Architecture

Publisher

ACM New York, NY, USA

Bibliometrics

Downloads (6 Weeks): 3, Downloads (12 Months): 56, Citation Count: 1

Additional Information:

abstract references cited by index terms collaborative colleagues

Tools and Actions:	Request Permissions Review this Article Save this Article to a Binder Display Formats: BibTeX EndNote ACM Ref

DOI Bookmark:	Use this link to bookmark this Article: http://doi.acm.org/10.1145/1378533.1378554 What is a DOI?

ABSTRACT

Although overlapping communication with computation is an important mechanism for achieving high performance in parallel programs, developing applications that actually achieve good overlap can be difficult. Existing approaches are typically based on manual or compiler-based transformations. This paper presents a pattern and library-based approach to optimizing collective communication in parallel high-performance applications, based on using non-blocking collective operations to enable overlapping of communication and computation. Common communication and computation patterns in iterative SPMD computations are used to motivate the transformations we present. Our approach provides the programmer with the capability to separately optimize communication and computation in an application, while automating the interaction between computation and communication to achieve maximum overlap. Performance results with a model application show more than a 90% decrease in communication overhead, resulting in 21% overall performance improvements.

REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

	1	Tarek S. Abdelrahman , Gary Liu, Overlap of computation and communication on shared-memory networks-of-workstations, Cluster computing, Nova Science Publishers, Inc., Commack, NY, 2001
	2	Ron Brightwell , Sue P. Goudy , Arun Rodrigues , Keith D. Underwood, Implications of application usage characteristics for collective communication offload, International Journal of High Performance Computing and Networking, v.4 n.3/4, p.104-116, August 2006 [doi>10.1504/IJHPCN.2006.010633]
	3	Anthony Danalis , Ki-Yong Kim , Lori Pollock , Martin Swany, Transformations to Parallel Codes for Communication-Computation Overlap, Proceedings of the 2005 ACM/IEEE conference on Supercomputing, p.58, November 12-18, 2005 [doi>10.1109/SC.2005.75]
	4	Sergei Gorlatch, Send-receive considered harmful: Myths and realities of message passing, ACM Transactions on Programming Languages and Systems (TOPLAS), v.26 n.1, p.47-56, January 2004 [doi>10.1145/963778.963780]
	5	Torsten Hoefler , Peter Gottschling , Andrew Lumsdaine , Wolfgang Rehm, Optimizing a conjugate gradient solver with non-blocking collective operations, Parallel Computing, v.33 n.9, p.624-633, September, 2007 [doi>10.1016/j.parco.2007.06.006]
	6	T. Hoefler, P. Kambadur, R. L. Graham, G. Shipman, and A. Lumsdaine. A Case for Standard Non-Blocking Collective Operations. In Recent Advances in Parallel Virtual Machine and Message Passing Interface, EuroPVM/MPI 2007, volume 4757, pages 125--134. Springer, 10 2007.
	7	Torsten Hoefler , Andrew Lumsdaine , Wolfgang Rehm, Implementation and performance analysis of non-blocking collective operations for MPI, Proceedings of the 2007 ACM/IEEE conference on Supercomputing, November 10-16, 2007, Reno, Nevada [doi>10.1145/1362622.1362692]
	8	Kamil Iskra, Pete Beckman, Kazutomo Yoshii, and Susan Coghlan. The influence of operating systems on the performance of collective operations at extreme scale. In Proceedings of Cluster Computing, 2006 IEEE International Conference, 2006.
	9	G. Liu and T.S. Abdelrahman. Computation communication overlap on network-of-workstation multiprocessors. In Proc. of the Int?l Conference on Parallel and Distributed Processing Techniques and Applications, pages 1635--1642, July 1998.
	10	Andrew Lumsdaine, Douglas Gregor, Bruce Hendrickson, and Jonathan Berry. Challenges in parallel graph processing. Parallel Processing Letters, 17(1):5--20, 2007 2007.
	11	Fabrizio Petrini , Darren J. Kerbyson , Scott Pakin, The Case of the Missing Supercomputer Performance: Achieving Optimal Performance on the 8,192 Processors of ASCI Q, Proceedings of the 2003 ACM/IEEE conference on Supercomputing, p.55, November 15-21, 2003
	12	Jose Carlos Sancho, Kevin J. Barker, Darren J. Kerbyson, and Kei Davis. Mpi tools and performance studies-quantifying the potential benefit of overlapping communication and computation in large-scale scientific applications. In Proceedings of the 2006 ACM/IEEE conference on Supercomputing, page 125. ACM Press, 2006.