PhD Viva Voice of Mr. Jasobanta Laha

The research focuses on the theoretical foundations and practical implications of dynamic load balancing in cloud computing environments. Cloud computing is a distributed computing paradigm where computational resources such as processing power, memory, and storage are delivered over the internet. In such environments, workloads often fluctuate dynamically, leading to uneven distribution of tasks among virtual machines (VMs). This imbalance may cause certain machines to become overloaded while others remain underutilized, resulting in increased response time and reduced system performance.

The concept of load balancing plays a critical role in maintaining system stability and efficiency. Load balancing refers to the process of distributing workloads evenly across multiple computing resources to optimize resource utilization, maximize throughput, minimize response time, and avoid overload conditions. The research proposes a novel dynamic approach, which continuously monitors system parameters and redistributes tasks among virtual machines in real time. Unlike static load balancing techniques, which follow predefined allocation rules, the dynamic approach adapts to changing workloads and system conditions.

The theoretical framework of the proposed model is based on task scheduling algorithms, resource monitoring mechanisms, and decision-making strategies that consider CPU utilization, memory usage, queue length, and execution time. By analyzing these parameters, the system intelligently assigns multiple jobs to multiple virtual machines. This ensures that the load on any single machine is reduced and that tasks are executed efficiently. The model emphasizes scalability, adaptability, and fault tolerance as essential components of modern cloud infrastructures.

A key theoretical contribution of the research lies in improving performance metrics such as makespan, throughput, and response time. By distributing tasks dynamically, the proposed approach minimizes idle time of virtual machines and prevents bottlenecks in the system. The balanced allocation of jobs enhances parallel processing capability, leading to faster program execution and improved Quality of Service (QoS). The approach also supports elastic cloud environments, where resources can be scaled up or down according to demand.