A generalized framework for energy savings in real-time multiprocessor systems

A generalized dynamic energy performance scaling (DEPS) framework is proposed for exploring application-specific energy-saving potential in multiprocessor systems. This software-centric framework takes advantage of possible power control mechanisms to trade off performance for energy savings. Three existing technologies, i.e., dynamic hardware resource configuration (DHRC), dynamic voltage frequency scaling (DVFS), and dynamic power management (DPM), have been employed in this framework to achieve the maximal energy savings. The problem of determining the optimal task allocation and DEPS configurations is formulated as an integer linear programming (ILP) problem. Several practical issues such as how to reduce measurement and computation time and how to reduce the configuration overhead are also addressed. The effectiveness of DEPS is validated through a case study.