A Strategy of Joint Service Placement and Request Dispatching for LEO Satellite MEC Networks

Low Earth Orbit (LEO) satellite Multi-Access Edge computing (MEC) is expected to realize the ubiquitous coverage for global users and provide low-delay MEC service. In LEO satellite MEC networks, although recent studies have solved the problem of requests dispatched to the appropriate edge servers, they mainly focus on the effective use of computing resources and ignore the need to pre-store a large amount of data to provide services. Due to the limited storage and computing resources of satellites, it is impossible to place all services and meet the requests of all users. Considering the significant correlation between service placement and request dispatching, we investigate the joint optimization of service placement and request dispatching in LEO satellite MEC networks, which is formulated as a Mixed Integer Linear Programming (MILP) problem. In order to avoid overloading of some satellites while other satellites are not fully utilized, the load balancing factor is introduced in this paper. Our goal is to improve the user served ratio, minimize the cost of request dispatching and improve the load balancing among satellites while ensuring the constraints of computation and storage resources. The simulation results verify that, compared with the baseline mechanism, the scheme of joint service placement and request scheduling with loading balancing (SPRDLB) that we proposed has better performance.