异构边缘网络中基于服务器休眠的协同计算策略

在异构边缘网络中,基站和边缘服务器的密集部署使系统功耗激增,为移动运营商带来了高额成本.同时,由于异构系统中任务的多样性和复杂性,高效的资源分配和任务迁移构成了严峻的挑战.针对上述问题,建立了服务器间的任务迁移和服务器睡眠模型,提出了一种面向异构边缘网络中服务器多睡眠模式的任务迁移算法.该算法先将问题按时隙划分为多个子问题并分别建模为混合整数非线性规划问题,再使用序列二次规划算法对每个子问题进行迁移决策求解,最后通过睡眠决策算法确定服务器的睡眠模式.仿真结果表示,所提算法不仅可以均衡边缘网络中非均匀分布的任务量,提高服务器资源的利用率,而且可以通过服务器多休眠状态显著降低系统能耗,提高系统性能.

Collaborative computing algorithm based on server sleep in heterogeneous edge networks

In a heterogeneous edge network, the dense deployment of base stations and edge servers has caused a surge in power consumption, and huge power consumption has become a major cost for mobile operators. At the same time, due to the diversity and complexity of tasks in heterogeneous systems, efficient resource allocation and task migration pose severe challenges. In response to the above problems, this thesis establishes a task migration between servers and server sleep model, and proposes a task migrate algorithm for multiple sleep modes of servers in heterogeneous edge networks. The algorithm divides the system into multiple sub-problems according to time slots and models them as integer nonlinear programming problems. Each sub-problem uses the sequential quadratic programming algorithm to solve the migration decision, and the sleep mode of the server is determined by the sleep decision algorithm. The simulation results show that the proposed algorithm can balance the unevenly distributed business volume in the edge network, improve the utilization rate of computing resources in the network, significantly reduce the energy consumption of the system through multiple server sleep states, and eventually improve system''s performance.