A Resource Allocation Method for Cloudlet Placement Based on PSO in Mobile Edge Computing
Haodong Liang, Jin Li, Yuan Rong · 2022 7th International Conference on Intelligent Computing and Signal Processing (ICSP) · 2022
Mobile Edge Computing was proposed to push data centers towards network edges for reducing the network latency of delivering cloud services to mobile devices. Cloudlet is one type of edge servers with one-hop network distance from mobile users, which can provide computing resources. However, due to the limited computing resources and high placement cost of Cloudlet, the distribution and mobility of devices in a region to provide computing services must be considered to allocate resources properly. In combination with the two-stage Cloudlet framework, this paper proposes a resource allocation method based on PSO (RAMP) to achieve edge cloud collaboration. This proposed method initializes and assigns values based on the Cloudlet location and coverage of Cloudlet placement, simulates the task movement process through each iteration of the PSO algorithm, completes allocation decisions and task offloading scheduling in a short time according to real-time task information and resource load, to make Cloudlet services available to as many users as possible and reduce the response time for each user. Experimental results show that the proposed resource allocation method can process the task offloading with shorter total time more stably and select a better resource allocation scheme with a smaller difference in algorithm running time.