Integrated Multi-dimensional Prioritization and Adaptive Transmission for Function Scheduling in Serverless Edge Computing

Serverless computing is a transformative technology that simplifies the management of underlying infrastructure. Deploying serverless computing in edge environment enhances application execution efficiency compared to centralized cloud. In serverless edge computing, efficient scheduling is critical for optimizing the performance of application execution. Applications are typically modeled as Directed Acyclic Graphs (DAGs), where the complex dependencies between functions and the distinct characteristics of each function present challenges to scheduling. Moreover, the constraints of edge resources further complicate scheduling, impeding the enhancement of Quality of Service (QoS) for users. In this paper, we propose an effective function scheduling strategy aimed at improving the performance of applications execution. Specifically, we achieve this by considering the topological structure of the DAG, the dependencies between functions, and the resource requirements. Based on these insights, we develop a new method for function prioritization and propose a corresponding function deployment strategy. Furthermore, we design a data transmission method to reduce latency during data transmission between functions with dependencies. Experimental evaluation using Alibaba Cluster Trace Data shows that the proposed method achieves superior performance in terms of cost-efficiency compared to existing algorithms.