Answers are grounded in this paper and cite their sources.
This paper investigates the issue of data distribution and caching in heterogeneous edge servers with different and limited storage and computing resources to maximize the overall quality of service for multiple users. We innovatively considers the relationship between data storage resource requirements, computing resource requirements, data sampling rate, and service quality in the issue of heterogeneous edge data distribution caching. In the research motivation, we discusses the caching pressure on heterogeneous edge servers and devices caused by data with different storage and computing resource requirements, as well as data caching with multiple sampling rate versions. In the system model and problem formalization, we proposes the optimization goal of maximizing the overall perceived quality of experience(QoE) of user groups, and formalizes it as a NP hard Knapsack problem. Based on proving the monotonic submodular property of the problem, this paper proposes a heterogeneous edge data distribution caching algorithm based on greedy strategy, and proves its performance to have an approximate ratio of (1−1/e)/2. Simulation experiments based on real datasets show that our algorithm can effectively optimize the overall user perception QoE under edge caching architecture, and has significant advantages compared to existing algorithms.
No takes yet. Share an insight, caveat, or question.
Li et al. (2023) studied this question.