A survey on the scheduling mechanisms in serverless computing: a taxonomy, challenges, and trends

A survey on the scheduling mechanisms in serverless computing: a taxonomy, challenges, and trends

18 February 2024 | Mohsen Ghorbian, Mostafa Ghobaei-Arani, Leila Esmaeili
This paper provides a comprehensive survey of scheduling mechanisms in serverless computing, addressing the challenges and trends in this emerging field. Serverless computing, characterized by a pay-as-you-go model and microservice architecture, is particularly suitable for IoT devices due to its resource-efficient nature. The scheduling algorithms in serverless computing are crucial for optimizing resource utilization, but they face challenges such as dynamic behavior, heterogeneous resources, and varying workloads. The paper proposes a taxonomy of six main scheduling approaches: Energy-aware, Data-aware, Deadline-aware, Package-aware, Resource-aware, and Hybrid. It discusses open issues and research challenges, emphasizing the need for efficient resource management in serverless environments. The contributions of the paper include a systematic review of existing studies, an evaluation of various scheduling techniques, and the identification of future research directions. The organization of the paper covers background, related surveys, research methods, technique classifications, discussions, open issues, and conclusions.This paper provides a comprehensive survey of scheduling mechanisms in serverless computing, addressing the challenges and trends in this emerging field. Serverless computing, characterized by a pay-as-you-go model and microservice architecture, is particularly suitable for IoT devices due to its resource-efficient nature. The scheduling algorithms in serverless computing are crucial for optimizing resource utilization, but they face challenges such as dynamic behavior, heterogeneous resources, and varying workloads. The paper proposes a taxonomy of six main scheduling approaches: Energy-aware, Data-aware, Deadline-aware, Package-aware, Resource-aware, and Hybrid. It discusses open issues and research challenges, emphasizing the need for efficient resource management in serverless environments. The contributions of the paper include a systematic review of existing studies, an evaluation of various scheduling techniques, and the identification of future research directions. The organization of the paper covers background, related surveys, research methods, technique classifications, discussions, open issues, and conclusions.
Reach us at info@study.space