Data Placement Strategies for Data-Intensive Computing over Edge Clouds
Xinliang Wei, Arrasy Rahman, Yu Wang · 2021
Edge computing has become an increasingly popular computing paradigm. Deploying edge clouds allows performing data-intensive computing at the edge of the network instead of a remote cloud to reduce data access latency and improve data processing efficiency. One of the key challenges in data-intensive edge computing is how to effectively place the data at the edge clouds such that the access latency to the data is minimized. In this paper, we study such a data placement problem in edge computing where different data items have diverse popularity. We first propose a data popularity based placement method when the data requests are unknown. It maps both data items and edge servers to a virtual plane and places data based on its virtual coordinate in the plane. We consider data popularity during both the mapping of data items to the plane and making the placement decision. We further propose an optimization-based placement strategy for the case when the data requests are known. By formulating an integer programming problem, our proposed solution aims to find the optimal placement decision. Simulation results show that both proposed strategies efficiently reduce the average latency of data access.