Energy use is currently an important issue in mobile embedded systems, and this will be continue in the future because of battery problems. We focused on a Dynamic Voltage Frequency Scaling (DVFS) algorithm to save as much energy as possible, and investigated the correlation of operations between memory and CPU. We introduce an Operational Intensity (OI)-based DVFS algorithm to follow the optimal frequency to conserve energy in embedded systems. Our proposed method was verified using FPMark of EEMBC, and conserved energy up to 17% in a memory-bound application.
A "roofline model" is a system performance and optimization guide for programmers and system engineers to apply in the design of future architectures. We review a conventional roofline model and propose a working set size (WSS)-based roofline model. Because of the recent increase in the performance of embedded systems, we investigated embedded systems based on an ARM architecture. Our proposed scheme presents practical guidelines for a multi-core architecture when the program uses a small WSS.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.