Implications of hierarchical N-body methods for multiprocessor architecture

We first examine the key architectural implications of realistically scaling a representative member of this important class of applications. Using scaling methods that reflect the concerns of an applications scientist leads to different conclusions than does naive scaling in terms of data set size. In particular, we show that under the most realistic scaling model, both the communications to computation ratio and the amount of cache memory per processor required for effective performance increase with scaling. We then examine the effect of a shared address space versus message passing as the communications abstraction. We show that lack of a shared address space substantially increase the programming complexity and performance overheads of a message-passing implementation.