A Network of Workstations (NOW) Approach to Spatial Data Analysis: The Case of Distributed Parallel Interpolation

Parallel processing has been widely demonstrated to be of significant value in reducing computation times for a range of geographical problems. Most papers have reported on the use of exotic and difficult-to-access dedicated parallel computing systems. However, researchers have begun to develop ways of exploiting collections of readily available workstations to implement off-the-shelf parallel computer systems. In this paper we report on the use of a Network of Workstations (NOW) environment that is applied to a typical spatial analysis problem- inverse distance weighted interpolation. Our results indicate that while a collection of networked workstations can be used to reduce run times from minutes to a few seconds, the implementation that we explored was not able to make fully efficient use of the available computing resources.