Hybrid Algorithm for Network Load Balancing Using Machine Learning

In modern computing environments characterized by high variability and complex workloads, traditional load balancing algorithms such as Round Robin and Least Connections are often lower in effectively distributing tasks and maintaining optimal performance. This paper presents a Hybrid- Machine Learning (Hybrid-ML) load balancing algorithm that combines the strengths of Fastest Response Load Balancing (FRLB) and Priority-based Load Balancing (PBLB) with advanced machine learning (ML) techniques. The Hybrid-ML algorithm leverages real-time data to predict optimal server allocations, dynamically adjusting priorities based on current server capacities, loads, response times, and task characteristics. This adaptive approach minimizes response times and ensures balanced load distribution across heterogeneous server environments. Recent literature emphasizes the growing necessity for intelligent load balancing solutions capable of adapting to dynamic conditions in cloud computing, IoT, and large-scale web services. By integrating ML, the Hybrid-ML algorithm provides a robust solution that learns from historical data and current system states to make informed decisions, enhancing overall system efficiency and reliability. Simulation results demonstrate that the Hybrid-ML algorithm outperforms traditional methods, achieving tangible reduction in average response time compared to FRLB and PBLB. This paper presents a comprehensive approach to load balancing that combines predictive analytics with traditional algorithms, offering an efficient solution for current computing challenges. The findings highlight the critical role of ML-driven hybrid models in advancing intelligent load balancing strategies.