Bharti Vijay Nikose
Department of Computer Engineering, Ramrao Adik Institute of Technology, D. Y. Patil Deemed to be University, Navi Mumbai, Maharashtra, India.

Gautam Murlidhar Borkar
Department of Information Technology, Ramrao Adik Institute of Technology, D. Y. Patil Deemed to be University, Navi Mumbai, Maharashtra, India.

DOI https://doi.org/10.33889/IJMEMS.2026.11.5.090

  ;

Abstract

Task scheduling in clouds is a multimodal task that involves integration of task-requirement analysis, Virtual Machine (VM) capacity analysis, dependency analysis, deadline analysis, and compatibility evaluation between task & VMs. Efficient task-scheduling models aim at maximizing cloud utilization, while maintaining higher deadline hit ratio, lower cloud effort, and higher task diversity levels. Most of the existing task scheduling models either, do not consider inter-task dependency while scheduling, or are highly complex which limits their scalability performance when applied to large-scale task sets. To overcome these issues, this study proposes design of a dependency-aware clustering model for improving task-scheduling performance via bioinspired optimizations. The proposed model initially scans input tasks and uses a Service Level Agreement for dependency resolutions. These dependency-resolved tasks are clustered on the basis of a score metric that combines make-span and deadline requirements. Clustering is facilitated via a combination of K-Means, Fuzzy C-Means, and Hierarchical Clustering techniques. The clusters along with VM configurations are processed by a Crow Search Optimization Model, which assists in mapping VMs to relevant tasks. The mapping process is facilitated via a deadline-aware & task-capacity-aware fitness function, which assists in improving its scheduling performance under heterogeneous task types. It is observed that the proposed model is showing improvement in achieving less scheduling delay, higher deadline hit ratio, increase in scheduling efficiency and reduction in energy required for scheduling tasks.

Keywords- Task scheduling, Crow search optimization, Service level agreement, Multicloud.

Citation

Nikose, B. V. & Borkar, G. M (2026). Enhancement of Task Scheduling Performance with Dependency Aware Clustering Model Using Bioinspired Optimizations in Multicloud. International Journal of Mathematical, Engineering and Management Sciences, 11(5), 2248-2267. https://doi.org/10.33889/IJMEMS.2026.11.5.090.