Decarbonisation of Data Centre Networks through Computing Power Migration
Zhigang Du, Hangji Yin, Xi Zhang, Heng Hu, T. Liu, Meng Hou, Spyros Giannelos, Goran Štrbac
Abstract
With the rapid development of Internet technologies and the widespread application of large-scale models, the computational load demand of data centers (DCs) continues to increase, exacerbating issues such as high energy consumption and operational costs. To address these challenges, DCs can improve their power and computing resource adjustment capabilities by migrating computational loads or incorporating energy storage systems to reduce operational costs. This paper aims to explore the economic impact of these two approaches on DCs. First, a computing power scheduling model for DCs is constructed. Then, we consider three different cases and use Yalmip and Cplex to solve the data migration strategy. Finally, the results demonstrate that the combination of energy storage and computing power migration not only facilitates the integration of renewable energy, but also yields better economic benefits.