Title: Dynamic resource allocation and scaling for adaptive systems in cloud environments

Authors: H. Annaiah; A. Rajesh

Addresses: Department of Computer Science and Engineering, JAIN Deemed-to-be-University, Bengaluru, Karnataka 560041, India; Department of Computer Science and Engineering, Government Engineering College, Krishnarajpete, Karnataka 571426, India ' Department of Computer Science and Engineering, JAIN Deemed-to-be University, Bengaluru, Karnataka 560041, India

Abstract: Efficient resource allocation, task prioritisation, and adaptive scaling are critical for modern cloud computing to handle dynamic workloads and optimise performance. Balancing resource demands, energy consumption, and timely task completion remains challenging. This paper proposes a novel integrated methodology combining four techniques tailored for cloud environments. Priority-based agglomerative hierarchical clustering categorises tasks based on urgency and resource requirements, enabling intelligent scheduling. Task prioritisation with round-robin scheduling ensures high-priority tasks are processed first while avoiding bottlenecks. Dynamic resource allocation leverages the wombat optimisation algorithm (WOA) for global exploration and the dung beetle optimiser (DBO) for local fine-tuning, maintaining solution diversity while minimising energy consumption. Reinforcement learning provides adaptive on-demand system scaling, optimising performance and cost in real time. The proposed approach significantly improves task execution efficiency, reduces energy usage, enhances scalability, and ensures timely task completion compared to traditional cloud computing methods.

Keywords: resource allocation; optimisation; workload; priority-based.

DOI: 10.1504/IJCC.2026.154441

International Journal of Cloud Computing, 2026 Vol.15 No.2, pp.206 - 225

Received: 17 Jun 2025
Accepted: 18 Aug 2025
Published online: 29 Jun 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article