Answers are grounded in this paper and cite their sources.
This model improves throughput by 19% and cuts energy consumption by 26% in cloud computing, indicating effective load balancing.
Objectives: To improve the throughput with minimum energy consumption in cloud load balancing by using proposed Position Sensitive Transformer Learning and Parallel Particle Swarm Optimization (PSTL-PPSO).Methods: This work proposes Position Sensitive Transformer Learning and Parallel Particle Swarm Optimization (PSTL-PPSO). Position Sensitive Attention-based Transformer Learning is initially applied to distribute enormous volume of data, to name a few being CPU usage, memory usage, network usage and potentially user-specific information. The contribution of PSTL-PPSO improving throughput, overhead efficiency with minimum response time. The collected data is then fed as input into a Position Sensitive Attention-based Transformer Learning model to analyze the relationships and patterns to comprehend prevailing system state. Based on the Position Sensitive Attention function the Transformer Learning model predicts optimal workload distribution across servers aiming to minimize makespan and energy consumption. Next, Parallel Particle Swarm Optimization (PPSO) algorithm is applied for fine-tuning cloud load balancing where a fitness function is provided to measure quality of each load balancing configuration on the basis of response time and throughput. The PPSO algorithm fine-tunes the position of each particle in an iterative fashion, exploring distinct load balancing alternatives in parallel fashion, therefore ensuring faster convergence optimally. Simulations are performed to validate the proposed PSTL-PPSO method in Cloud Simulator and Java programming language in terms of makespan time, energy consumption, high accuracy, associated overhead and throughput. Findings: PSTL-PPSO method improves the throughput by 19% and reduces makespan 17%, energy consumption by 26%, overhead efficiency by 26%. Novelty: The integration of Position Sensitive Attention-based Transformer Learning with Parallel Particle Swarm Optimization to enable adaptive, energy-efficient, and fast-converging cloud load balancing under dynamic workloads. This hybrid model is applicable to cloud platforms, data centers, and edge computing environments, where efficient workload distribution and resource utilization are critical.
No takes yet. Share an insight, caveat, or question.
Sinduja et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: