1Research Scholar, Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India
2Professor, Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, Andhra Pradesh, India
3Professor, Department of IT, S.R.K.R. Engineering College, Bhimavaram, Andhra Pradesh, India
Cloud computing systems usually have to deal with a vast number of tasks to be executed on distributed resources. One key challenge in such systems is how to allocate resources and schedule cloudlet effectively among many tasks that come simultaneously. The quality of a cloud platform's performance is dependent on how well it is able to allocate these jobs to its virtual machines. Good schedule results in balanced workload, better responsiveness of the system, and multiple cloudlets could run concurrently without waiting for each other. Scheduling tasks is the process of associating user requests or application tasks to virtual machines in a manner that enhances execution and resource utilization. To minimize total completion time, enhance system throughput and avoid resource congestion, a good scheduling strategy should be developed. But this task is hard to compute. Combinatorial scheduling is usually thought to be NP-hard because the space of candidate solutions can increase exponentially with the number of tasks and computing resources. As tasks (n), for execution, are to be assigned to (m) resources the number of different possible schedules rapidly increases rendering the identification of the best schedule hardly feasible in limited time. To deal with above challenge, an efficient scalable scheduling framework is proposed in cloud. The proposed framework clusters the virtual machines based on their workload and behavior. This categorization step contributes to further reshape the resource pool, and also confines the searching space for scheduling. Once VMs are clustered, a Q-ant colony optimization method is used to find suitable allocations among tasks, VMs, and physical servers. The process of optimization simulates cooperative search behavior such as ant colonies, which the system can discover efficient task allocation routes gradually. With iterating over evaluation and adjustment, the algorithm refines its scheduling choices to achieve better resource balance in the cloud environment. Simulation results show that our proposed model can enhance the efficiency of the cloud environment. The complex results show that about 6% decrease in computational overhead and energy consumption of the cloud compared with the state-of-the-scheduling algorithm. But at the same time, the framework preserves necessary quality of service properties that the cloud users are accustomed to, since the completed tasks will have acceptable response times and the resource usage over infrastructure is relatively balanced.
Keywords: Quality of service, cluster based task scheduling, cloud servers, cloud scheduling.
How to cite this article: Rupasri M, Varma GPS, Indukuri H. A Dynamic Task Scheduling Model for Efficient Resource Management in Large-Scale Cloud Environments. Int J Drug Deliv Technol. 2026;16(6s): 615-627; DOI: 10.25258/ijddt.16.6s.87
Source of support: None
Conflict of interest: None