sponsors
usenix conference policies
Managing Green Datacenters Powered by Hybrid Renewable Energy Systems
Chao Li, University of Florida; Rui Wang, Beihang University; Tao Li, University of Florida; Depei Qian, Beihang University; Jingling Yuan, Wuhan University of Technology
The rapidly growing server energy expenditure and the warning of climate change have forced the IT industry to look at datacenters powered by renewable energy. Existing proposals on this issue yield sub-optimal performance as they typically assume certain specific type of renewable energy sources and overlook the benefits of cross-source coordination. This paper takes the first step toward exploring green datacenters powered by hybrid renewable energy systems that include baseload power supply, intermittent power supply, and backup energy storage. We propose GreenWorks, a power management framework for green high-performance computing datacenters powered by renewable energy mix. Specifically, GreenWorks features a hierarchical power coordination scheme tailored to the timing and capacity of different renewable energy sources. Using real datacenter workload traces and renewable power generation data, we show that our scheme allows green datacenters to achieve better design trade-offs.
Open Access Media
USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.
author = {Chao Li and Rui Wang and Tao Li and Depei Qian and Jingling Yuan},
title = {Managing Green Datacenters Powered by Hybrid Renewable Energy Systems},
booktitle = {11th International Conference on Autonomic Computing (ICAC 14)},
year = {2014},
isbn = {978-1-931971-11-9},
address = {Philadelphia, PA},
pages = {261--272},
url = {https://www.usenix.org/conference/icac14/technical-sessions/presentation/li_chao},
publisher = {USENIX Association},
month = jun
}
connect with us