Energy Efficient Dynamic Computing Resource Allocation in Centralized Radio Access Networks

Abstract

Computing resource allocation algorithms are very important in centralized radio access networks to reduce the power consumption and construction cost of base stations. Most of the current research cannot make the best use of computing resources because they take whole processing tasks of one BS as the allocation object which is too large for a processor. Therefore, an Energy Aware Dynamic Resource Allocation (EADRA) algorithm is proposed in this paper, which takes a whole BS or the sub-task of a BS as the allocation object dynamically to achieve the target that the processing ability of computing resources can be fully utilized. Simulation results show that the EADRA effectively decreases the number of allocated processor cores when compared with existing algorithms. In addition, EADRA reduces the power consumption more and more significantly compared to that of the existing algorithms when the number of Virtual BSs (VBSs) is gradually increased.