Pipelining and overlapping for MPI collective operations

Proceedings - Conference on Local Computer Networks, LCN

Volumn 2003-January, Issue , 2003, Pages 548-557

  Worringen, J ^a  

^a NEC LABORATORIES EUROPE   (Germany)

Author keywords

Analytical models; Europe; Laboratories; Message passing; Multiprocessor interconnection networks; National electric code; Operating systems; Parallel programming; Performance evaluation; Pipeline processing

Indexed keywords

ANALYTICAL MODELS; COMPUTER NETWORKS; COMPUTER OPERATING SYSTEMS; INTERCONNECTION NETWORKS (CIRCUIT SWITCHING); LABORATORIES; PARALLEL PROCESSING SYSTEMS; PARALLEL PROGRAMMING; PIPELINE CODES; PIPELINE PROCESSING SYSTEMS; PIPELINES;

COMMUNICATION ALGORITHMS; EUROPE; MESSAGE PASSING INTERFACE; NATIONAL ELECTRIC CODE; PERFORMANCE EVALUATION; PIPELINE PROCESSING; POINT-TO-POINT MESSAGE PASSING; SCALABLE COHERENT INTERFACE;

MESSAGE PASSING;

EID: 84944415126     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/LCN.2003.1243181     Document Type: Conference Paper

References (23)

1. R. Bhoedjang, T. Rühl, and H. E. Bal. Efficient Multicast on Myrinet using Link-Level Flow Control. In Proceedings of the 1998 International Conference on Parallel Processing, pages 381-390, Minneapolis, August 1998.
2. Dolphin Interconnect Solutions. SISCI Interface Specification, 2.1.1 edition, May 1999.
3. S. Fleischmann and D. Golan. MPI Collective Communication on the Convex Exemplar SPP-1000 Series Scalable Parallel Computer. Presentation at MPI Developers Conference (MPIDC95), 1995.
4. W. Gropp, E. Lusk, N. Doss, and A. Skjellum. A high-performance, portable implementation of the MPI message passing interface standard. Parallel Computing, 22(6):789-828, Sept. 1996.
5. L. P. Huse. Collective Communication on Dedicated Clusters of Workstations. In J. D. et.al., editor, Recent Advances in PVM and MPI, number 1697 in LNCS, pages 469-476, September 1999.
6. L. P. Huse, K. Omang, H. Bugge, H. Ry, A. Haugsdal, and E. Rustad. SCI: Scalable Coherent Interface. Architecture software for high-performance compute clusters, chapter ScaMPI - Design and Implementation. Number 1734 in LNCS. Springer, 1999.
7. IEEE, editor. Standard for Scalable Coherent Interface (SCI). Number 1596 in IEEE Standards. The Institute of Electrical and Electronics Engineers, Inc., 1992.
8. M. C. Liaanen and H. Kohmann. SCI: Architecture and Software for High Peformance Compute Clusters, chapter Dolphin SCI Adapter Cards, pages 71-86. Number 1734 in LNCS. Springer, 1999.
9. G. R. Luecke, B. Raffin, and J. J. Coyle. The Performance of MPI Collective Communication Routines for Large Messages on the Cray T3E-600, the Cray Origin 2000, and the IBM SP. Journal of Performance Evaluation and Modelling for Computer Systems, July 1999.
10. Message-Passing Interface Forum. MPI: A Message-Passing Interface Standard . http://www.mpi-forum.org, 1995.
11. Message-Passing Interface Forum. MPI-2: Extentsions to the Message-Passing Interface. http://www.mpi-forum.org, 1997.
12. P. Mitra, D. G. Payne, L. Shuler, R. van de Geijn, and J. Watts. Fast Collective Communication Libraries, Please. Technical Report TR-95-22, Department of Computer Sciences, The University of Texas at Austin, Austin, Texas 78712-1188, June 1995.
13. S. Oral and A. George. Multicast performance analysis for high-speed torus networks. In 27th Annual IEEE Conference on on Local Computer Networks (LCN), Workshop on High Speed Local Networks, pages 619-628. IEEE Computer Society, November 2002.
14. Pallas GmbH. Pallas MPI Benchmark 2.0. http://www.pallas.com, 2002.
15. F. Petrini, S. Coll, E. Frachtenberg, and A. Hoisie. Hardware- and Software-Based Collective Communication on the Quadrics Network. In IEEE International Symposium on Network Computing and Applications 2001 (NCA 2001), Boston, MA, February 2002.
16. R. Rabenseifner. A new optimized MPI reduce algorithm. High-Performance Computing-Center, November 1997. http://www.hlrs.de/mpi/myreduce.html.
17. P. Sanders and J. F. Sibeyn. A Bandwidth Latency Tradeoff for Broadcast and Reduction. In A. Bode, T. Ludwig, W. Karl, and R. Wismüller, editors, Proceedings of 6th International EuroPar Conference, number 1900 in LNCS, Munich, August 2000.
18. F. Seiffert, J. Worringen, and W. Rehm. Using Arbitrary Memory Regions for SCI Communication. In SCI Europe Conference, Dublin, October 2001. Trinity College.
19. SISCI Consortium. Standard Software Infrastructures for SCI-based Parallel Systems (SISCI). http://www.parallab.uib.no/projects/sisci, August 1997.
20. R. Thakur and W. Gropp. Improving the Performance of Collective Operations in MPICH. Submitted to publication for Euro-PVM/MPI 2003, May 2003.
21. S. S. Vadhiyar, G. E. Fagg, and J. Dongarra. Automatically Tuned Collective Communications. In J. E. Donnelley, editor, Proceedings of Supercomputing 2000, Dallas, November 2000.
22. J. Worringen and T. Bemmerl. MPICH for SCI-connected Clusters. In SCI Europe Conference, Toulouse, September 1999. Conference Stream of EuroPar Conference.
23. J. Worringen, F. Seiffert, and T. Bemmerl. Efficient Asynchronous Message Passing via SCI with Zero-Copying. In SCI Europe Conference, Dublin, October 2001. Trinity College.