AMBITION: Ambient Temperature Aware VM Allocation for Energy Efficient Edge Data Centers
TimeWednesday, July 13th6pm - 7pm PDT
LocationLevel 2 Lobby
DescriptionIn edge data centers, servers can have various ambient temperature. Ambient temperature has a large impact on the on-chip temperature of CPUs, affecting the performance of servers. However, the conventional virtual machine (VM) allocation techniques do not consider the impact of ambient temperature, leading to inefficient decisions. In this paper, we propose an ambient temperature aware VM allocation technique, called AMBITION. In our experiments on our edge data centers, AMBITION shows the speedup of 50.3% on average (up to 61.2%), compared to the conventional VM allocation techniques, while saving system-wide energy consumption by 5.9% on average (by up to 13.6%).