Improved task scheduling in heterogeneous distributed systems using intelligent greedy harris hawk optimization algorithm

In the context of heterogeneous computer systems, the utilization of task scheduling algorithms is crucial for enhancing system performance. A major challenge in the development of efficient scheduling algorithms for such systems lies in reducing execution time. However, achieving this objective is not a straightforward process. Due to the complexity of these problems, exact algorithms are not feasible for solving them, primarily because of the vast search space involved. Consequently, approximate algorithms become necessary. Among these, meta-heuristic algorithms are notable for their ability to explore large search spaces and generate suitable solutions. Nevertheless, these algorithms are not flawless as they may become trapped in local optima and failing to scan the entire sample space. To address task scheduling problems in heterogeneous computer systems, this research proposes the Intelligent Harris Hawk Optimization algorithm. The algorithm was employed to tackle the task scheduling problem in heterogeneous systems, while also considering common issues associated with such algorithms, including early convergence and limited exploration capability during the exploration phase. To overcome these challenges, the exploration phase was separated from the exploitation phase. This approach aimed to ensure that the algorithm explores the entire problem space without necessarily achieving optimality. However, if the algorithm becomes trapped in a local optimum during the exploitation phase, it may not be able to escape and achieve early convergence. To address this issue, a solution called Bigmax to Fitmax was proposed for escaping local optima in the soft besiege phase, while a unique technique called Crisscross Mutation was introduced for the hard besiege phase. In the final stage, the proposed algorithm will be compared with several other algorithms, including HHO (Harris Hawk Optimization), MOSCOA (A multi-objectives scheduling algorithm based on cuckoo optimization), HACG-TS (Hybrid approach based on cuckoo optimization algorithm and genetic algorithm), GA (Genetic Algorithm), HEFT (Heterogeneous Earliest-Finish-Time), and CPOP (Critical-Path-on-a-Processor), to solve the scheduling problem in a heterogeneous environment.