作者: Zhonghao Sun , Fanxin Kong , Xue Liu , Xingshe Zhou , Xi Chen
DOI: 10.1109/IGCC.2014.7039141
关键词: Engineering 、 Data center 、 Electrical grid 、 Schedule 、 Electricity 、 Workload 、 Real-time computing 、 Electric vehicle 、 Cost efficiency 、 Power (physics)
摘要: A data center is designed to have the capacity matching spike workload from a geographical region. This design concept comes with significant waste of expense on power peak charged by electrical grid, since fluctuations cause large variations in demand. On other hand, electric vehicles (EVs) been emerging as major electricity consumers due their demand for battery charging. In this paper, we propose ValleyFill method that explores EVs fill valleys geographically distributed centers without increasing peaks. Leveraging diversities processing and temporal flexibilities EV charging, determines routing scheme charging schedule improve cost efficiency charge. We evaluate proposed real-world traces arrival patterns. The result shows our significantly improves saves up 6% total bills. observe larger gap between its valley leads less time migration.