Stochastic programming on cost optimization in geographically distributed Internet data centers

Abstract

The development of cloud computing has accelerated the growth of internet data centers (IDCs) worldwide. While a large portion of the energy consumption aroused by intense computation introduces great operation expenditure to the IDC enterprises. To manage the overall cost and utilize resources to its full extent, this paper paid attention on the spatio-temporal workload allocation among the geographically distributed IDCs within a cloud, with the guarantee of the workload completion time and the consideration of computing service delay penalty by introducing the cost of inconvenience. Apart from the effort of the workload migration, the spatio-temporal variance of the renewable energies in the data center microgrids (DMGs) is fully considered in this paper. What's more, as the power consumed by the IDCs are mostly converted into heat, waste heat recovery process is embedded in each IDC to demonstrate the effectiveness of the repurposed heat, which would be used by the residential heating demand in the thermal system, on total cost reduction and energy usage efficiency of the whole operating system. Applying real-life data traces of electricity price, renewable energies and heating demand, the extensive evaluations demonstrate that both spatial and temporal complementary effort on supply side and demand side, power and thermal complementary effort could significantly reduce the overall cost for the IDC enterprise.