Component Infrastructure for Managing Performance Data and Runtime Adaptation of Parallel Applications

Abstract. Component-based software engineering (CBSE) has been gain-ing popularity in parallel scientific computing, facilitating the creation and management of large multidisciplinary, multideveloper application codes and providing opportunities for improved performance and numer-ical accuracy. The CBSE approach enables automation of traditionally manual application configuration and tuning tasks. In addition to en-capsulating data and functions, components enable the encapsulation of nonfunctional properties (metadata) whose values affect the performance of a component. By first automatically collecting performance data and associated metadata, and then analyzing it to classify different meth-ods, we can automate the selection of the best-performing component instance among many functionally equivalent implementations. To sup-port this adaptivity, we have designed a set of interfaces and components for managing databases for performance and system information, anal-ysis, and runtime control. We describe our initial implementations of these interfaces and discuss their use in selecting linear solvers in inex-act Newton methods for nonlinear partial differential equations (PDEs) and dynamic selection of the most appropriate partitioning strategy at runtime in combustion applications.