Brasil

Abstract

Brasil is a self-contained service which can be deployed across a cluster to provide a dataflow workload distribution and communication aggregation mechanism. Together with our dataflow shell, named PUSH, it is intended to be used for the management of non-traditional super computing applications as well as provide a mechanism to manage in-situ analysis and vizualization of more traditional high performance computing simulations. This paper describes our experiences implementing and deploying a prototype of Brasil on a BlueGene/P supercomputer.