基于epsilon-greedy的边缘服务器选择模型

如今,5G的时代已经到来,万物互联成为可能。在这种情况下,移动通信技术在人们日常生活和社会发展中的地位进一步突出。用户本地的计算卸载到边缘服务器中,从而解决用户设备在计算性能、存储等方面的不足。一般来说,一个用户周围会存在多个边缘服务器,由此便引发了边缘服务器的选择问题。文章重点介绍了在5G的背景下基于epsilon-greedy的边缘服务器选择问题,以及多臂老虎机模型、epsilon-greedy算法,多臂老虎机模型实现边缘服务器选择,对比了随机选择和epsilon-greedy的优劣。

Edge server selection model based on epsilon-greedy

Now, the era of 5 G has come and the Internet of everything is possible. In this case, mobile communication technology in People’s Daily life and social development in a further prominent position. The local computation of the user is unloaded to the edge server, so as to solve the shortage of the user device in computing performance, storage and so on. In general, there are multiple edge servers around a user, which raises the issue of edge server selection. This paper focuses on the edge-server selection problem based on epsilongreedy in the context of 5 G, introduces the multi-arm slot machine model, the epsilon-greedy algorithm, and the multi-arm slot machine model to achieve edge server selection, and compares the advantages and disadvantages of random selection and epsilon-greedy, as well as the comparison of different epsilon.