2016/01/12 by Umer Liqat, Zorana Banković, Liqat, Umer +5
Computer Science · Engineering · #Distributed #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Green IT and Sustainability #Parallel #Parallel Computing and Optimization Techniques #Programming Languages (cs.PL) #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.1601.02800
openalex publication_date 2016/01/12 · openalex created_date 2022/09/27 · openalex updated_date 2026/07/28
The ever increasing number and complexity of energy-bound devices (such as\nthe ones used in Internet of Things applications, smart phones, and mission\ncritical systems) pose an important challenge on techniques to optimize their\nenergy consumption and to verify that they will perform their function within\nthe available energy budget. In this work we address this challenge from the\nsoftware point of view and propose a novel parametric approach to estimating\ntight bounds on the energy consumed by program executions that are practical\nfor their application to energy verification and optimization. Our approach\ndivides a program into basic (branchless) blocks and estimates the maximal and\nminimal energy consumption for each block using an evolutionary algorithm. Then\nit combines the obtained values according to the program control flow, using\nstatic analysis, to infer functions that give both upper and lower bounds on\nthe energy consumption of the whole program and its procedures as functions on\ninput data sizes. We have tested our approach on (C-like) embedded programs\nrunning on the XMOS hardware platform. However, our method is general enough to\nbe applied to other microprocessor architectures and programming languages. The\nbounds obtained by our prototype implementation can be tight while remaining on\nthe safe side of budgets in practice, as shown by our experimental evaluation.\n