Performance Modeling of Extreme-Scale Systems and Applications

Abstract

In this talk we will describe novel methodologies for performance analysis, modeling and prediction applicable to extreme-scale parallel architectures and applications.  We will present ongoing and planned performance projects in PAL at Los Alamos National Laboratory, based on the new methods that can be applied to the current extreme-scale and next-generation parallel systems.  In addition to the methodology for performance modeling, a variety of applications of modeling will be presented, including architectural design, workload characterization, code optimization and system diagnostics.

We will conclude by summarizing a number of factors that in our view will significantly impact the development and performance of future generation parallel systems.