A Hybrid AI-Driven Swarm Approach for Optimized Task Scheduling in Cloud Computing Environment

The increasing number of internet users turning to modern cloud computing in the coming years presents challenges, especially when considering constraints such as limited completion times, optimal resource utilization, low operational costs, and short task scheduling durations. Task scheduling becomes more complex as the number of tasks and available resources grow, leading to a rapid increase in potential solutions and combinations, making the problem harder to solve. To address this, the paper proposes an Enhanced Task Scheduling (ETS) model driven by optimized artificial intelligence, incorporating swarm-based methods in cloud computing environments. The proposed method uses an Artificial Neural Network (ANN) with swarm-based optimization, specifically Moth Flame Optimization (MFO), to balance scheduling by classifying Virtual Machines (VMs) based on CPU utilization, memory, and bandwidth. Initially, tasks are scheduled conventionally, then the ETS model is optimized using ANN with the Back-Propagation Algorithm (ANN-BPA) to address resource allocation and balancing challenges. The study also introduces an alternative model using only ANN to evaluate the ETS model's efficiency, focusing on task allocation, completion, execution time, and energy consumption. The ANN-BP A-based model outperforms existing techniques and simple ANN models, increasing resource utilization by 7.54% and reducing task execution time by 0.6 seconds.