A Novel Hybrid MSA-CSA Algorithm for Cloud Computing Task Scheduling Problems

• Published in: Symmetry (Basel) Vol. 15; no. 10; p. 1931
• Main Authors: Alsubai, Shtwai, Garg, Harish, Alqahtani, Abdullah
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01-10-2023
• Subjects: Access control; advanced encryption standard; Algorithms; Allocations; Analysis; Bandwidths; Cloud computing; Cybersecurity; Efficiency; hybrid model; moth swarm algorithm; Network latency; Network management systems; Optimization algorithms; Performance enhancement; Performance evaluation; Quality of service; Resource scheduling; Resource utilization; Scheduling; security; Security management; Swarm intelligence; Task scheduling; Workloads
• ISSN: 2073-8994; 2073-8994
• DOI: 10.3390/sym15101931

Recently, the dynamic distribution of resources and task scheduling has played a critical role in cloud computing to achieve maximum storage and performance. The allocation of computational tasks in the cloud is a complicated process that can be affected by some factors, such as available network bandwidth, makespan, and cost considerations. However, these allocations are always non-symmetric. Therefore, it is crucial to optimize available bandwidth for efficient cloud computing task scheduling. In this research, a novel swarm-based task scheduling with a security approach is proposed to optimize the distribution of tasks using available resources and encode cloud information during task scheduling. It can combine the Moth Swarm Algorithm (MSA) with the Chameleon Swarm Algorithm (CSA) for the task scheduling process and utilizes the Polymorphic Advanced Encryption Standard (P-AES) for information security of cloud scheduled tasks. The approach offers a new perspective for utilizing swarm intelligence algorithms to optimize cloud task scheduling. The integration of MSA and CSA with P-AES enables the approach to provide efficient and secure task scheduling by exploiting the strengths of used algorithms. The study evaluates the performance of the proposed approach in terms of the degree of imbalance, makespan, resource utilization, cost, average waiting time, response time, throughput, latency, execution time, speed, and bandwidth utilization. The simulation is carried out using a wide range of tasks from 1000 to 5000. The results show that the approach provides an innovative solution to the challenges of task scheduling in cloud environments and improves the performance of cloud services in terms of effectiveness and security measures.