Multi-objective scheduling technique based on hybrid hitchcock bird algorithm and fuzzy signature in cloud computing

As the number of requests in the cloud increases, providers face more problems (e.g., task scheduling and resource management). The exhaustive search for determining the optimal solutions of scheduling problem is usually impractical and hence metaheuristic algorithms have been widely used to evolve solutions for task scheduling. In this paper, we firstly propose a hybrid meta-heuristic algorithm called Hybrid Fuzzy Hitchcock Bird (HFHB) and second, a multi-objective form of HFHB (MOHFHB) is introduced to solve multi-objective problems (e.g., task scheduling). The HFHB algorithm consists of three main enhancements: (a) The first random population of birds is improved, (b) The attack regulator parameter is set with a fuzzy Sugenosignature, and (c) The dead birds are replaced with new birds. In multi-objective form (MOHFHB), two concepts (i.e., crowding distance and ranking non-dominated solution) are added to determine the optimal Pareto front. In the first part, HFHB and MOHFHB are evaluated as two global optimizers. In the second part, HFHB and MOHFHB are evaluated for a task scheduling problem against Moth Search Algorithm with DE (MSDE), Enhanced Multi-Verse Optimizer (EMVO), Fuzzy Modified Particle Swarm Optimization (FMPSO), and Simulated-annealing-based Bees Algorithm (SBA). The results indicate that HFHB improves makespan by 12.57%, 38.61%, and 35.75%, and resource utilization by 1.14%, 7.30%, and 5.25% compared to FMPSO, MSDE, and EMVO, respectively.

Automated summary

This study introduces a hybrid metaheuristic algorithm called HFHB for task scheduling problems, which combines fuzzy features and optimization algorithms to achieve significant progress in solving multi-objective problems. The algorithm demonstrates better performance compared to other algorithms in experimental evaluations.