Answers are grounded in this paper and cite their sources.
Background/Objectives: The rapid growth of the internet and cloud-based applications has led the world to adopt cloud computing. As the number of users grow there rises some problems, virtual machine allocation and cloudlet scheduling are among them. Efficient resource allocation and scheduling are crucial for balancing the load on virtual machines, as they minimize delays and optimize performance. Method: This research work proposes a predictive mechanism using Triplet Attention and Quantum-behaved Particle Swarm Optimization (QPSO). By incorporating triplet Attention, VM allocation and load balancing are performed efficiently, and the QPSO optimizes the Scheduling. For predictive models’ performance evaluation, R2 score, MAE, and RMSE metrics are evaluated. Findings: The model achieved 96.7% R2 score and the error metrics are 0.34 MAE and 0.68 RMSE, for experimental outcomes of cloud environment load balancing score, execution time, resource utilization, energy usage, and throughput are evaluated. The experimental results show that average execution time is reduced to 6.34ms from 9.38ms (AFPA), average energy consumption is reduced to 21.18Kwh from 29.95KWh (AFPA), load balancing improved to 98.06 from 98 (AFPA) throughput is increased from 51.63 (FSAOS) to 83.33 cloudlet/ms and load balancing score improved from 98.16 (AFPA) to 98.26. Novelty: The proposed model ensembles deep learning (Triplet Attention) with optimization (QPSO), and optimizes multiple key metrics at same time, the proposed model optimizes total six key metrics where the other studies only optimize two or three metrics. This work provides a comprehensive model for VM allocation and cloudlet scheduling.
No takes yet. Share an insight, caveat, or question.
Singh et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: