This paper aims at maximizing the profit associated with runninggeographically dispersed green data centers, which offer multiple classes ofservice. To this end, we formulate an optimization framework which relies onthe accuracy of the G/D/1 queue in characterizing the workload distribution,and taps on the merits of the workload decomposition into green and brownworkload served by green and brown energy resources. Moreover, we take intoaccount of not only the Service Level Agreements (SLAs) between the datacenters and clients but also different deregulated electricity markets of datacenters located at different regions. We prove the convexity of ouroptimization problem and the performance of the proposed workload distributionstrategy is evaluated via simulations.
展开▼
机译：本文旨在最大程度地利用与运行地理位置分散的绿色数据中心相关的利润，这些数据中心提供多种服务。为此，我们制定了一个优化框架，该框架基于G / D / 1队列的准确性来表征工作负载分布，并利用工作负载分解为绿色和棕色能源所服务的绿色和棕色工作负载的优点。此外，我们不仅要考虑数据中心和客户之间的服务水平协议（SLA），还要考虑到位于不同地区的数据中心的不同管制市场的电力市场。我们证明了优化问题的凸性，并且通过仿真评估了所提出的工作负载分配策略的性能。
展开▼
