Additional author:

distributed memory message-passing parallel implementation of a finite-volume discretization of the primitive equations in the Community Atmosphere Model is presented. Due to the data dependencies resulting from the polar singularity of the latitude-longitude coordinate system, we employ two separate domain decompositions within the dynamical core – one in latitude/level space, and the other in longitude/latitude space. This requires that the data be periodically redistributed between these two decompositions. In addition, the domains contain halo regions that cover the nearest neighbor data dependencies. A combination of several techniques, such as one-sided communication and multithreading, are presented to optimize data movements. The resulting algorithm is shown to scale to very large machine configurations, even for relatively coarse resolutions. 4Introduction Atmospheric general circulation models (AGCMs) are key tools for weather prediction and climate research. They also require large computing resources: even the largest current supercomputers cannot keep pace with the desired increases in the resolution of these models. AGCMs consist, roughly speaking, of the dynamics, which calculates the atmospheric flow, and the physics, i