EVALUATION OF 2 PROGRAMMING PARADIGMS FOR HETEROGENEOUS COMPUTING

In this paper, we evaluate two different programming paradigms for het erogeneous computing, Cluster-M and Heterogeneous Associative Computin g (HAsC). These paradigms can efficiently support heterogeneous networ ks by preserving a level of abstraction without containing any archite ctural details. The paradigms are architecturally independent and scal able for various network and problem sizes. Cluster-M can be applied t o both coarse-grained and fine-grained networks. Cluster-M provides an environment for porting heterogeneous tasks onto the machines in a he terogeneous suite such that resource utilization is maximized and the overall execution time is minimized. HAsC models a heterogeneous netwo rk as a coarse-grained associative computer. It is designed to optimiz e the execution of problems where the program size is small compared w ith the amount of data processed. Unlike other existing heterogeneous orchestration tools which are MIMD based, HAsC is for data-parallel SI MD associative computing. Ease of programming and execution speed are the primary goals of HAsC. We evaluate how these two paradigms can be used together to provide an efficient scheme for heterogeneous program ming. Finally, their scalability issues are discussed. (C) 1993 Academ ic Press, Inc.