Abstract

In embedded systems, performance and power are important inter-related issues that cannot be decoupled. Expensive and extensive simulations in a processor design space are usually required to verify whether a design meets both performance and power requirements. In this paper, an analytical co-optimization approach based on an integrated workload, performance and power model for modern processors is described and studied. A design space consisting of more than 15 architectural and workload parameters can be quickly explored for co-optimization. Validation with measured results obtained from simulators as well as physical processors showed that the model has a good degree of accuracy. We shall describe the details of approach and the model, and show how to apply the approach to the problem of co-optimizing the power and performance of processor design. With the completeness, flexibility and efficiency, our approach provides clear insights into the tradeoffs of designs for performance and power.

Full Text
Paper version not known

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

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.