Chapter 8 A Programming Environment forNetwork

Heterogeneous associative computing (HAsC) is a new network heterogeneous computing (NHC) paradigm that is a combination of associative computing and heterogeneous computing as related to superconcurrency. The goal of this high performance computing environment is to increase the throughput of very large applications such as those on the scale of grand challenge problems. HAsC is the result of a number of years of research and development work at Kent State University. 8.1 INTRODUCTION The concept of heterogeneous associative computing came about from the desire to increase the throughput of computationally intensive application codes. The thought was that if a program has portions that are inherently well suited to a spe-ciic architecture, they should be executed on that architecture. Moreover, no single computer is capable of processing all of the capabilities of all machines. That is, only a heterogeneous network can combine multiple instruction multiple data (MIMD), single instruction multiple data (SIMD), and vector machines into a \single virtual machine." As promising as the \execute a code on its optimal machine" concept is, there were concerns over moving data from place to place, because network communication costs are extremely high in a widely distributed environment. Thus it was decided that the associative computing concept of broadcasting the instructions to the processors while making data sedentary could be used to reduce network traac. Another concern was the diiculty of writing programs for distributed heterogeneous applications. The complexity of programming is greatly compounded by 1