2014/12/19 by Julien Mineraud, Mineraud, Julien, Liang Wang +5
Computer Science · #Distributed #FOS: Computer and information sciences #Networking and Internet Architecture (cs.NI) #Parallel #Performance (cs.PF) #and Cluster Computing (cs.DC) #cs.DC #cs.NI #cs.PF
paper · pdf · doi:10.48550/arxiv.1412.6382
13 pages
arxiv created 2014/12/19 · arxiv updated 2014/12/22
The ICT industry today is placed as one of the major consumers of energy, where recent reports have also shown that the industry is a major contributor to global carbon emissions. While renewable energy-aware data centers have been proposed, these solutions have certain limitations. The primary limitation is due to the design of data centers which focus on large-size facilities located in selected locations. This paper addresses this problem, by utilizing in-network caching with each router having storage and being powered by renewable energy sources (wind and solar). Besides placing contents closer to end users, utilizing in-network caching could potentially increase probability of capturing renewable energy in diverse geographical locations. Our proposed solution is dual- layered: on the first layer a distributed gradient-based routing protocol is used to discover the paths along routers that are powered by the highest renewable energy, and on the second layer, a caching mechanism will pull the contents from the data centre and place them on routers of the paths that are discovered by our routing protocol. Through our experiments on a testbed utilizing real meteorological data, our proposed solution has demonstrated increased quantity of renewable energy consumption, while reducing the workload on the data centers.