XPS: FULL: CCA: Collaborative Research: CASH: Cost-aware Adaptation of Software and Hardware

点击翻译按钮获取中文摘要

Programming for and leveraging the benefits of scalable and highlyparallel computer architectures is becoming increasinglychallenging. In order to ease programmer and user effort in achievingefficient use of parallel architectures and to enable optimal usage ofparallel hardware both within a single chip and across a data center,this project explores the co-design of a fine-grain configurable (downto the ALU, FPU, Cache bank, fetch unit, etc) architecture along witha software optimizing runtime system which controls the architecture'sconfiguration. This runtime management system understands high-levelgoals and constraints such as optimizing for power, latency, cost, orcomplex mixed-goals and dynamically allocates fine-grain resources tomeet the constraints. This project will evaluate and design a configurablemanycore-inspired architecture and a self-adapting runtime system.This project will investigate the creation of a complete, scalable,self-adaptive computing system; and will push the boundaries ofadapting systems by utilizing hardware that is configurable andmonitorable below the processor core level. By providing such aflexible and highly monitored architecture to the adaptation runtime,this project will explore the scalability of adaptive runtimealgorithms to handle a game-changing increase in the number ofcontrollable parameters. This project will explore the extent towhich it is fruitful to configure a parallel architecture.