CoRaiS: Lightweight Real-Time Scheduler for Multiedge Cooperative Computing

Multiedge cooperative computing that combines constrained resources of multiple edges into a powerful resource pool has the potential to deliver great benefits, such as a tremendous computing power, improved response time, and more diversified services. However, the mass heterogeneous resources composition and lack of scheduling strategies make the modeling and cooperating of multiedge computing system particularly complicated. This article first proposes a system-level state evaluation model to shield the complex hardware configurations and redefine the different service capabilities at heterogeneous edges. Second, an integer linear programming model is designed to cater for optimally dispatching the distributed arriving requests. Finally, a learning-based lightweight real-time scheduler, CoRaiS is proposed. CoRaiS embeds the real-time states of the multiedge system and requests information, and combines the embeddings with a policy network to schedule the requests, so that the response time of all requests can be minimized. Evaluation results verify that the CoRaiS can make a high-quality scheduling decision in real-time, and can be generalized to other multiedge computing system, regardless of the system scales. Characteristic validation also demonstrates that the CoRaiS successfully learns to balance loads, perceive real-time state and recognize heterogeneity while scheduling.

I. Introduction

Edge computing brings computation and storage resources closer to the sources of data, facilitating the processing of the client data at the network periphery while meeting stringent response time requirements. Some great progresses have been achieved, especially in the mobile edge computing [1], [2]. However, practical applications often reveal challenges. As shown in Fig. 1, each edge hosts a diverse set of services, and it typically serves multiple clients. The distribution of clients among edges exhibits nonuniformity, with variations in both the number of requests submitted by each client and the specific service they require. This complexity can potentially degrade service quality, especially when an edge is inundated with an excessive number of client requests. Fig. 1.

Illustration of unbalanced workloads and resource utilization of edge computing.