SYMPOSIUM KEYNOTE PRESENTATIONS

ABSTRACT

An overview is given of the Navy's efforts in converting serial ocean models to massively parallel computing systems. Starting with the data parallel approach CM Fortran on the Connection Machines CM200 and CM5, continuing with HPF (High Performance Fortran) and MPI (Message Passing, including the Cray-specific highly efficient SHMEM routines) on both distributed memory (IBM SP2) and shared memory (Cray T3E, Origin2000) platforms. to current applications of OpenMP for clusters of shared memory processors, we review the Navy's development work in this area. We discuss the efficiency of parallel algorithms for the various component of the ocean model, including time marching, solving Helmholtz equations by iterative as well as direct solvers, and give timings on scalability and communications overhead.

Getting More Information

To obtain more information about the meeting send e-mail to: fscc98@pax.st.usm.edu.

Return to PAX
Return to USM

USM
inquire@delphi.st.usm.edu
Updated
http://pax.st.usm.edu