Enhanced Task Scheduling Algorithm Using Harris Hawks Optimization Algorithm for Cloud Computing

Abstract

Amongst the most transformational technologies nowadays, cloud computing can provide resources such as CPU, memory, and storage over secure internet connections. Due to its flexibility and resource availability with guaranteed QoS, cloud computing allows comprehensive business and research adoptions. Despite the rapid development, resource management remains one of the significant challenges, especially handling task scheduling efficiently in this environment. Task scheduling strategically assigns tasks to available resources so that Quality of Service (QoS) metrics are effectively related to response time and throughput. This paper proposes an Enhanced Harris Hawks Optimization (EHHO) algorithm for scheduling cloud tasks to mitigate the common limitations found in existing algorithms. EHHO integrates a dynamic random walk strategy, enhancing exploration capabilities to avoid premature convergence and significantly improving scalability and resource allocation efficiency. Simulation outcomes reveal that EHHO minimizes makespan by up to 75%, memory usage by up to 60%, execution time by up to 39%, and cost by up to 66% compared to state-of-the-art algorithms. These benefits demonstrate that EHHO can optimize resource allocation while being highly scalable and reliable. Consistent performance over various stacks such as Kafka, Spark, Flink, and Storm further evidences the superiority of EHHO in handling complex scheduling challenges in dynamic cloud computing environments.