AI-powered federated task scheduling and self-healing framework in dynamic cloud systems

Demirbaga, Umit, Rana, Omer ORCID: https://orcid.org/0000-0003-3597-2646, Anjum, Ashiq and Singh Aujla, Gagangeet 2025. AI-powered federated task scheduling and self-healing framework in dynamic cloud systems. Presented at: 17th International Conference on Utility and Cloud Computing (UCC), Sharjah, United Arab Emirates, 16-19 December 2024. 2024 IEEE/ACM 17th International Conference on Utility and Cloud Computing (UCC). IEEE, pp. 300-305. 10.1109/ucc63386.2024.00049

Full text not available from this repository.

Abstract

Federated cloud environments have emerged to integrate multiple cloud providers like AWS, Azure, and Google Cloud seamlessly into cloud computing. Optimising resource utilisation and ensuring high availability in such environments pose significant challenges. This paper comprehensively investigates federated task scheduling algorithms and self-healing mechanisms in autonomous federated cloud setups. The research objectives include the development of an independent task-scheduling algorithm capable of intelligently distributing computing tasks across federated clouds based on workload characteristics, resource availability, and network latency. Furthermore, the study investigates implementing self-healing mechanisms to detect faults and performance degradation, triggering automatic recovery processes for uninterrupted service availability. The proposed approaches are evaluated through real-world experiments, considering diverse cloud workloads and failure scenarios, focusing on resource utilisation efficiency, system performance, and the effectiveness of the self-healing mechanisms in mitigating cloud failures and maintaining seamless operations within the federated environment.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Schools > Computer Science & Informatics
Publisher:	IEEE
ISBN:	9798350367201
Last Modified:	07 May 2025 11:00
URI:	https://orca.cardiff.ac.uk/id/eprint/178094

Actions (repository staff only)

Edit Item