Characterizing and Predicting Program Behavior and its Variability

• Authors: Evelyn Duesterwald;Calin Cascaval;Sandhya Dwarkadas
• Venue: Proceedings of the 12th International Conference on Parallel Architectures and Compilation Techniques
• Year: 2003

To reach the next level of performance and energy efficiency, optimizations are increasingly applied in a dynamic and adaptive manner. Current adaptive systems are typicallyreactive and optimize hardware or software in response to detecting a shift in program behavior. We argue that program behavior variability requires adaptive systems to be predictive rather than reactive. In order to be effective, systems need to adapt according to future rather than most recent past behavior.In this paper we explore the potential of incorporating prediction into adaptive systems. We study the time-varying behavior of programs using metrics derived from hardware counters on two different micro-architectures. Our evaluation shows that programs do indeed exhibit significant behavior variation even at a granularity of millions of instructions. In addition, while the actual behavior across metrics may be different, periodicity in the behavior is shared across metrics. We exploit these characteristics in the design of on-line statistical and table-based predictors. We introduce a new class of predictors, cross-metric predictors, that use one metric to predict another, thus making possible an ef.cient coupling of multiple predictors. We evaluate these predictors on the SPECcpu2000 benchmark suite and show that table-based predictors outperform statistical predictors by as much as 69% on benchmarks with high variability.