Enabling flexible collective communication offload with triggered operations

Proceedings - Symposium on the High Performance Interconnects, Hot Interconnects

Volumn , Issue , 2011, Pages 35-42

  Underwood, Keith D ^a   Coffman, Jerrie ^a   Larsen, Roy ^a   Hemmert, K Scott ^b   Barrett, Brian W ^b   Brightwell, Ron ^b   Levenhagen, Michael ^b  

^a INTEL CORPORATION   (United States)

^b SANDIA NATIONAL LABORATORIES   (United States)

Author keywords

Allreduce; Collective; MPI; Offload

Indexed keywords

ALL-REDUCE; APPLICATION SCALABILITY; BUILDING BLOCKES; COLLECTIVE; COLLECTIVE COMMUNICATIONS; COLLECTIVE OPERATIONS; KEY COMPONENT; LOW LATENCY; MPI; OFFLOAD; PERFORMANCE IMPROVEMENTS;

ALGORITHMS; SEMANTICS;

COMMUNICATION;

EID: 80555144336     PISSN: 15504794     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/HOTI.2011.15     Document Type: Conference Paper

References (16)

1. A. Moody, J. Fernandez, F. Petrini, and D. K. Panda, "Scalable NIC-based reduction on large-scale clusters," in Proceedings of the ACM/IEEE SC2003 Conference, November 2003.
2. A. Wagner, D. Buntinas, R. Brightwell, and D. K. Panda, "Application-bypass reduction for large-scale clusters," in Proceedings 2003 IEEE Conference on Cluster Computing, December 2003.
3. R. E. Riesen, K. T. Pedretti, R. Brightwell, B. W. Barrett, K. D. Underwood, T. B. Hudson, and A. B. Maccabe, "The Portals 4.0 message passing interface," Sandia National Laboratories, Tech. Rep. SAND2008-2639, April 2008.
4. S. L. Scott and G. Thorson, "Optimized routing in the Cray T3D," in PCRCW '94: Proceedings of the First International Workshop on Parallel Computer Routing and Communication. London, UK: Springer-Verlag, 1994, pp. 281-294.
5. G. Almasi, G. Dozsa, C. C. Erway, and B. Steinmacher- Burow, "Efficient implementation of allreduce on BlueGene/L collective network," in Proceedings of EuroPVM/MPI, September 2005.
6. R. Sivaram, R. Kesavan, D. K. Panda, and C. Stunkel, "Where to provide support for efficient multicasting in irregular networks: Network interface or switch?" in Proceedings of the International Conference on Parallel Processing, August 1998, pp. 452-459.
7. D. Buntinas, D. K. Panda, and P. Sadayappan, "Fast NICbased barrier over Myrinet/GM," in Proceedings of the International Parallel and Distributed Processing Symposium, April 2001.
8. W. Yu, D. Buntinas, R. L. Graham, and D. K. Panda, "Efficient and scalable barrier over Quadrics and Myrinet with a new NIC-based collective message passing protocol," in Proceedings of the Workshop on Communication Architecture for Clusters, April 2004.
9. D. Buntinas, D. K. Panda, J. Duato, and P. Sadayappan, "Broadcast/multicast over Myrinet using NIC-assisted multidestination messages," in Proceedings of the Fourth International Workshop on Communication, Architecture, and Applications for Network-Based Parallel Computing, January 2000.
10. D. Buntinas and D. K. Panda, "NIC-based reduction in Myrinet clusters: Is it beneficial?" in Proceedings of the SAN- 02 Workshop (in conjunction with HPCA), February 2002.
11. D. Roweth and A. Pittman, "Optimised global reduction on QsNet-II," in Thirteenth IEEE Symposium on High- Performance Interconnects (HotI'05), August 2005.
12. A. Wagner, H.-W. Jin, D. K. Panda, and R. Riesen, "NICbased offload of dynamic user-defined modules for Myrinet clusters," in Proceedings of the 2004 IEEE International Conference on Cluster Computing, September 2004, pp. 205- 214. (Pubitemid 40822371)
13. R. Rabenseifner, "Optimization of collective reduction operations," in Computational Science - ICCS 2004, 2004.
14. K. D. Underwood, M. Levenhagen, and A. Rodrigues, "Simulating Red Storm: Challenges and successes in building a system simulation," in 21st International Parallel and Distributed Processing Symposium (IPDPS'07), March 2007.
15. B. W. Barrett and K. S. Hemmert, "An application based MPI message throughput benchmark," in Proceedings of the 2009 IEEE International Conference on Cluster Computing, September 2009.
16. K. B. Ferreira, P. Bridges, and R. Brightwell, "Characterizing application sensitivity to OS interference using kernellevel noise injection," in SC '08: Proceedings of the 2008 ACM/IEEE conference on Supercomputing. Piscataway, NJ, USA: IEEE Press, 2008, pp. 1-12.