Optimisation of tail latency in Edge computing

Abstract

Edge computing has emerged as a promising paradigm for meeting the requirements of resourceintensive applications by processing data at the network edge. The aim of our project is to optimise the end to end tail latency arising in Edge Computing. We propose a reinforced learning algorithm with Deep Q learning to achieve this . We test our approach on YAFS - a simulator for fog computing. We find that our approach outperforms all baseline approaches with sufficient training .