A OptiNet-Learning Model for Efficient Task Scheduling in Cloud Systems

Many organizations widely use cloud computing applications. Load balancing is a sub-domain of cloud computing that can effectively schedule tasks for high-demand applications. Cloud Service Providers (CSPs) primarily focus on providing on-demand, quality-of-service (QoS) without interruption. In cloud computing, effective task scheduling, resource allocation, and resource usage become more complex for large-scale data. There are many challenges in cloud-based applications due to high workloads, unpredictable attacks, and network delays. This paper presents OptiNet-Learning, a new Ensemble Optimization Learning (EOL) technique that addresses several issues discussed earlier. The proposed model mainly focused on the following. Firstly, dynamic resource allocation, reducing the time frames for dynamic allocation, and rapid task allocation. Secondly, deep reinforcement learning (DRL) refines results, applies N-iterations to detect missing tasks, refines existing (EOL) results, and demonstrates high performance. Researchers use the proposed algorithm to benchmark datasets and evaluate it quantitatively. Finally, the proposed approach shows superior performance across throughput and other metrics.