Optimization of Resource and Energy in Distributed Systems Using Unified Genetic Algorithm

Cloud and large distributed systems must ensure resource scheduling, energy management, and resource allocation. However, there exist complex and dynamic workloads, which may cause inefficient resource distribution, increased energy consumption, cost of operation and time delays which ultimately lead to reduced Quality of Experience (QoE). To address these issues the Unified Genetic Algorithm (UGA) is proposed, a proactive approach in optimization which helps achieve relatively better balance between CPU and memory usage across multiple nodes in a distributed system. UGA, was tested using the Materna workload trace and subjected to comparison with other existing load balancing algorithms such as Firefly, Coral Reef Optimization and Novel Family. It is found that UGA is superior with regard to efficiency in scheduling as it has revealed an improvement of 6.72% in average when compared to state of the art algorithms and proved to be beneficial in optimal resource allocation and improvement in system performance.