Dr. Jay Hoeflinger
Jay Hoeflinger received a B.S. in 1974, an M.S. in 1977, and a PhD in 1998, all in Computer Science and all from the University of Illinois at Urbana-Champaign. He worked as a scientific programmer for 8 years, supporting geological research. He spent 5 years in industry, working as a systems programmer, a systems analyst, and a manager. Then he was lured back to the University of Illinois to be a member of the Compiler Group at the Center for Supercomputing Research and Development starting in 1985, developing the Cedar Fortran compiler. He worked as a Senior Software Engineer doing compiler research and supporting the Polaris parallelizing compiler project for 6 years. In 1998, he began work as a Senior Research Scientist at the Center for Simulation of Advanced Rockets, an ASCI center at the University of Illinois.
Fortran 90 has become increasingly popular in the scientific community as a language for writing application codes, because (among many other things) it offers storage management facilities and more flexible data structures than does Fortran 77. To make it even more useful, the two major parallel programming paradigms have industry-standard Fortran 90 bindings. SPMD parallelism may be expressed through the use of MPI calls, and shared memory parallelism may be expressed through OpenMP. But the two paradigms cause certain Fortran 90 constructs to display different performance characteristics.
This talk will compare MPI with OpenMP in a general way, then describe an experiment run on the Origin 2000 in which an application using MPI was re-implemented with OpenMP in such a way that the performance effects of the paradigms on each Fortran 90 construct were comparable. The performance differences will be described in detail, and conclusions will be drawn about the trade-offs involved in writing high-performance parallel programs using Fortran 90 on the Origin 2000.
Everyone is invited and welcome to attend the seminars in this series.