Designing high performance and scalable MPI intra-node communication support for clusters

Proceedings - IEEE International Conference on Cluster Computing, ICCC

Volumn , Issue , 2006, Pages

  Chai, Lei ^a   Hartono, Albert ^a   Panda, Dhabaleswar K ^a  

^a Ohio State University

Author keywords

Cluster computing; Intra node communication; MPI; Multi core processor; Non Uniform Memory Access (NUMA)

Indexed keywords

CLUSTER COMPUTING; CLUSTER ENVIRONMENTS; DATA LOCALITY; DUAL CORES; HIGH PERFORMANCE COMPUTING; INTERNATIONAL CONFERENCES; INTRA-NODE COMMUNICATION; INTRA-NODE LATENCY; L2 CACHE; LARGE SYSTEM SIZE; MEMORY ARCHITECTURES; MEMORY SPACES; MPI; MULTI-CORE PROCESSOR; NEW DESIGN; NON-UNIFORM MEMORY ACCESS (NUMA); NUMA SYSTEMS; OPTIMAL PERFORMANCES; OPTIMIZATION STRATEGIES;

CHLORINE COMPOUNDS; DATA TRANSFER; HIGH PERFORMANCE LIQUID CHROMATOGRAPHY; MACHINE DESIGN; SHAPE MEMORY EFFECT; TELECOMMUNICATION SYSTEMS;

COMPUTER SYSTEMS;

EID: 46049106540     PISSN: 15525244     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/CLUSTR.2006.311850     Document Type: Conference Paper

References (18)

1. http://lse.sourceforge.net/numa/faq/.
2. http://www.amd.com/.
3. http://www.intel.com/.
4. http://valgrind.org/.
5. MPI over InfiniBand Project. http://nowlab.cse.ohio-state.edu/projects/ mpi-iba/.
6. MPICH-MX Software, http://www.myri.com/scs/download-mpichmx.html.
7. MPICH2. http://www.mcs.anl.gov/mpi/.
8. Multicore Processor Technology. http://www.dell.com/downloads/global/ power/ps2q05-20050103-Fruehe.pdf.
9. Darius Buntinas, Guillaume Mercier, and William Gropp. Data Transfers Between Processes in an SMP System: Performance Study and Application to MPI. In International Conference on Parallel Processing, 2006.
10. Darius Buntinas, Guillaume Mercier, and William Gropp. The design and evaluation of Nemesis, a scalable low-latency message-passing communication subsystem. In International Symposium on Cluster Computing and the Grid, 2006.
11. Lei Chai, Sayantan Sur, Hyun Wook Jin, and D. K. Panda. Analysis of Design Considerations for Optimizing Multi-Channel MPI over InfiniBand. In Workshop on Communication Architecture on Clusters, 2005.
12. Per Ekman and Philip Mucci. Design Considerations for Shared Memory MPI Implementations on Linux NUMA Systems: An MPICH/MPICH2 Case Study. http://www.cs.utk.edu/ mucci/latest/pubs/AMD-MPI-05.pdf.
13. Edgar Gabriel, Graham E. Fagg, George Bosilca, Thara Angskun, Jack J. Dongarra, Jeffrey M. Squyres, Vishal Sahay, Prabhanjan Kambadur, Brian Barrett, Andrew Lumsdaine, Ralph H. Castain, David J.Daniel, Richard L. Graham, and Timothy S. Woodall. Open MPI: Goals, Concept, and Design of a Next Generation MPI. Euro PVM/MPI, September 2004.
14. W. Gropp, E. Lusk, N. Doss, and A. Skjellum. A High-Performance, Portable Implementation of the MPI, Message Passing Interface Standard. In Parallel Computing, 2006.
15. H. W. Jin, S. Sur, L. Chai, and D. K. Panda. Design and Performance Evaluation of LiMIC (Linux Kernel Module for MPI Intra-node Communication) on InfiniBand Cluster. In International Conference on Parallel Processing, 2005.
16. Myricom Inc. Portable MPI Model Implementation over GM, March 2004.
17. Jeffrey M. Squyres and Andrew Lumsdaine. A Component Architecture for LAM/MPI. In Proceedings, 10th European PVM/MPI Users' Group Meeting, number 2840 in Lecture Notes in Computer Science, Venice, Italy, September / October 2003. Springer-Verlag.
18. Sun Microsystems Inc. Memory Placement Optimization (MPO). www.opensolaris.org/os/community/performance/mpo_overview.pdf.