Abstract
This paper describes a unique visualization tool that has been used to analyze performance of the Cplant(tm) clusters at Sandia National Laboratories.As commodity cluster systems grow in size and complexity, understanding performance issues becomes more and more difficult.We have developed a tool that facilitates visual performance analysis within the context of the physical and runtime environment of a system. Combining an abstract system model with color-coding for both performance and job information enables quick fault isolation as well as insight into complex system behavior.