Multivariate geographic clustering using a Beowulf-style parallel computer

Abstract The authors present an application of multivariate non-hierarchical statistical cluster-ing to geographic environmental data from the 48 conterminous United States in order to produce maps of regions of ecological similarity called ecore-gions. Nine input variables thought to aect the growth of vegetation are clustered at a resolution of one square kilometer. These data represent over 7.8 million map cells in a 9-dimensional data space. For the analysis, the authors built a 126-node heterogeneous cluster|aptly named the Stone SouperComputer|out of surplus PCs. The authors developed a parallel iterative statistical clustering algorithm which uses the MPI message passing rou-tines, employs a classical master/slave single pro-gram multiple data (SPMD) organization, performs dynamic load balancing, and provides fault toler-ance. In addition to being run on the Stone Souper-Computer, the parallel algorithm was tested on other parallel platforms without code modication. Finally, the results of the geographic clustering are presented