This paper discusses optimization issues to be addressed in producing a major improvement in the performance of a state-of-the-art general circulation model of the atmosphere on massively parallel computer systems. Our work is based on the UCLA Atmospheric General Circulation Model (AGCM). The UCLA AGCM is a
grid-point model
, which uses a three dimensional staggered grid to compute flow variables around the globe. The current goal is to produce a code that runs at the 50 GFLOP level by early 1998.
We focus on load balancing strategies and single node optimization techniques. The target architecture for the single node optimization is the CRAY T3D. In the near future, the work will be extended to the CRAY T3E. This research is been carried out in the framework of NASA High Performance Computing and Communication program (HPCC).