Multi-attribute-based self-stabilizing algorithm for leader election in distributed systems

Abstract

Cloud computing, edge computing, replicated databases, and various modern technologies adopt distributed computing concepts for a reliable, high-performance, large-scale computational platform. In distributed computing, leader election is a fundamental problem because the elected leader helps coordinate and utilize the resources efficiently. Several state-of-the-art works mentioned that a good quality leader is essential as it improves system performance, simplifies management procedures, reduces coordinational complexity, and makes the system more fault-tolerant. However, designing a self-stabilizing leader election algorithm with weak assumptions in a failure-prone distributed environment is challenging. This paper proposes a multi-attribute-based, self-stabilizing, network partition-tolerant leader election method for failure-prone distributed systems. Here, based on the system requirements, the pertinent attributes of a good quality leader are identified and assigned weights according to their importance. Then, the identified attributes and their weight are used to elect a suitable node as the leader. We show that the algorithm is self-stabilizing and can tolerate multiple nodes and link failures. Further, we analyze the proposed algorithm’s time, communication, and bit complexities. We consider a distributed database system scenario to simulate the proposed method and compare it with existing approaches to evaluate and validate the proposed method’s performance and the elected leader’s quality. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.