Open Issues in MPI Implementation
|Title||Open Issues in MPI Implementation|
|Publication Type||Conference Paper|
|Year of Publication||2007|
|Authors||Thakur, R, Gropp, WD|
|Conference Name||12th Asia-Pacific Computer Systems Architecture Conference (ACSAC 2007)|
|Conference Location||Seoul, Korea|
MPI (the Message Passing Interface) continues to be the dominant programming model for parallel machines of all sizes, from small Linux clusters to the largest parallel supercomputers such as IBM Blue Gene/L and Cray XT3. Although the MPI standard was released more than 10 years ago and a number of implementations of MPI are available from both vendors and research groups, there are many areas in which MPI implementations still need improvement. In this paper, we discuss several such areas, including performance, scalability, fault tolerance, support for debugging and verification, topology awareness, collective communication, derived datatypes, and parallel I/O. We also present results from some experiments with several MPI implementations (MPICH2, Open MPI, Sun, IBM) on a number of platforms (Linux clusters, Sun, and IBM SMPs) that demonstrate the need for performance improvement in one-sided communication and support for multithreaded programs.