Abstract

A growing number of processors have CPU cores that implement the same instruction set architecture (ISA) using different microarchitectures. The underlying motivation for single-ISA heterogeneity is that a diverse set of cores can enable runtime flexibility . Modern processors are subject to strict power budgets, and heterogeneity provides the runtime scheduler with more latitude to decide the level of performance a program should have based on the amount of power that can be spent. We argue that selecting a diverse set of heterogeneous cores to enable flexible operation at runtime is a non-trivial problem due to diversity in program behavior. We further show that common evaluation methods lead to false conclusions about diversity. Finally, we suggest the KS statistical test as an evaluation metric. The KS test is the first step toward a heterogeneous design methodology that optimizes for runtime flexibility.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call