MPI on the I-WAY: a wide-area, multimethod implementation of the Message Passing Interface

High-speed wide-area networks enable innovative applications that integrate geographically distributed computing, database, graphics, and networking resources. The Message Passing Interface (MPI) can be used as a portable, high-performance programming model for such systems. However, the wide-area environment introduces challenging problems for the MPI implementor, because of the heterogeneity of both the underlying physical infrastructure and the authentication and software environment at different sites. We describe an MPI implementation that incorporates solutions to these problems. This implementation, which was developed for the I-WAY distributed-computing experiment, was constructed by layering MPICH on the Nexus multithreaded runtime system. Nexus provides automatic configuration mechanisms that can be used to select and configure authentication, process creation, and communication mechanisms in heterogeneous systems.

[1]  Matthew Haines,et al.  ROPES: SUPPORT FOR COLLECTIVE OPERATIONS AMONG DISTRIBUTED THREADS , 1995 .

[2]  K. Mani Chandy,et al.  CC++: A Declarative Concurrent Object Oriented Programming Notation , 1993 .

[3]  Rick Stevens,et al.  Sharing visualization experiences among remote virtual environments , 1995 .

[4]  John C. Hart,et al.  The CAVE: audio visual experience automatic virtual environment , 1992, CACM.

[5]  Warren Smith,et al.  Software infrastructure for the I-WAY high-performance distributed computing experiment , 1996, Proceedings of 5th IEEE International Symposium on High Performance Distributed Computing.

[6]  Ian T. Foster,et al.  Overview of the I-Way: Wide-Area Visual Supercomputing , 1996, Int. J. High Perform. Comput. Appl..

[7]  Carl Kesselman,et al.  Generalized communicators in the Message Passing Interface , 1996, Proceedings. Second MPI Developer's Conference.

[8]  Anthony Skjellum,et al.  Extending the message passing interface (MPI) , 1994, Proceedings Scalable Parallel Libraries Conference.

[9]  Ian T. Foster,et al.  The Nexus Approach to Integrating Multithreading and Communication , 1996, J. Parallel Distributed Comput..

[10]  Anthony Skjellum,et al.  A High-Performance, Portable Implementation of the MPI Message Passing Interface Standard , 1996, Parallel Comput..

[11]  John Shalf,et al.  Galaxies Collide On the I-Way: an Example of Heterogeneous Wide-Area Collaborative Supercomputing , 1996, Int. J. High Perform. Comput. Appl..

[12]  J. Geisler,et al.  Multimethod Communication for High-Performance Metacomputing Applications , 1996, Proceedings of the 1996 ACM/IEEE Conference on Supercomputing.

[13]  William Gropp,et al.  Skjellum using mpi: portable parallel programming with the message-passing interface , 1994 .

[14]  Alan F. Blackwell,et al.  Programming , 1973, CSC '73.

[15]  William Gropp,et al.  Mpich working note: creating a new mpich device using the channel interface , 1995 .

[16]  KesselmanCarl,et al.  Near-Real-Time Satellite Image Processing , 1996 .

[17]  Carl Kesselman,et al.  Near-real-time satellite image processing: metacomputing in CC++ , 1996, IEEE Computer Graphics and Applications.

[18]  K. Mani Chandy,et al.  Fortran M: A Language for Modular Parallel Programming , 1995, J. Parallel Distributed Comput..

[19]  Gul Agha,et al.  Research directions in concurrent object-oriented programming , 1993 .