A Discrete Interval-Based Multi-Objective Memetic Algorithm for Scheduling Workflow With Uncertainty in Cloud Environment

IEEE Transactions on Network and Service Management, Journal Year: 2022, Volume and Issue: 20(3), P. 3020 - 3037

Published: Nov. 24, 2022

In consideration of the uncertainty scientific workflows, an interval-based multi-objective cloud workflow scheduling problem is investigated, which widely exists in environment. This aims at allocating to public The uncertain workload and communication data as well processing ability bandwidth resources are represented by interval number, models these variables. objectives minimize total execution time cost. To address this problem, a discrete memetic algorithm (DIMOMA) proposed. A hybrid initial strategy employed generate potential population. With contribution-based selection mechanism, self-adaptive genetic operators designed perform global search space. Then, novel local procedure incorporated intensification accelerate convergence. comprehensive computational experiment comparisons with several meta-heuristics adapted from related problems conducted based on extended benchmark set. simulated results reveal that proposed method can achieve better trade-off fronts between cost workflow. On performance metric hypervolume measures both cost, DIMOMA improve 3.90%, 9.30%, 6.25%, 7.74% compared EMS-C, MOACS, ch-PICEA-g, I_MaOPSO, respectively. Besides, robustness, means difference lower upper bounds solutions obtained over smaller than state-of-the-art algorithms.

Language: Английский

---

An Improved Glowworm Swarm Optimization Based on Various Mutation Operators

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 106359 - 106384

Published: Jan. 1, 2024

Glowworm Swarm Optimization (GSO) is a population-based optimization algorithm that successfully solves numerous problems. Nonetheless, the convergence speed required to reach optimal solutions can be made more efficient by skipping local optima. Also, considerable attention tuning parameter of crucial improve speed. In this study, three variants GSO are proposed using various mutation operators (Gaussian, Cauchy, and Lévy) its prevent it from getting stuck in optimum. The small random changes provided Gaussian help fine-tuning position Glowworms. Meanwhile, Cauchy offers large assist movement operator exploring wide area search space. Lévy characterized occasional jumps, which have potential explore problem space effectively. performance accuracy methods studied based on famous multimodal unimodal benchmark test functions, as well CEC2014 suite. Additionally, we experimented with set engineering effects settings improved discussed Response Surface Methodology (RSM). Results revealed suggested algorithms, offer better than basic other variants. comparison state-of-the-art GGSO obtained best results for 68.75%, 63.33% functions CEC2014, respectively. statistical tests show superiority over modified algorithms.

Language: Английский

---

GEMM, a Genetic Engineering-Based Mutual Model for Resource Allocation of Grid Computing

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 128537 - 128548

Published: Jan. 1, 2023

Resource sharing, selection, and aggregation are vital functions of grid computing. However, managing resources in a grid-based environment is stimulating task. It necessary to update the topographical dispersal possessed by various organisations with usage guidelines, financial frameworks, load, availability patterns. Users servers have different objectives, methods, needs. This article suggests cost-effective framework for resource management computing look at address these difficulties. The proposed has three main functions, which help construction, load balancing, allocation. A Genetic engineering approach been implemented establish relationship between pool jobs nodes that improve utilization. methodology also optimizes overall cost minimizing turnaround time. results research compared commonly used algorithms claiming 1.5 10% better results.

Language: Английский

---

GCRS – Green Computing Resource Scheduler: An Optimized Energy Efficient Cloud Data Centers to Scale Down Carbon Emission

Published: Oct. 18, 2023

Cloud computing is a powerful technology that rapidly growing in popularity with energy consumption being major concern. data centers consume lot of energy, which can lead to high operating costs and carbon emissions. This significant bottleneck restricting the development cloud computing. If people continue rely on traditional centers, environmental impact will only get worse. Green solutions are needed address this challenge. designed utilize resources while minimizing efficiently. providers offer different levels service, defined service-level agreements (SLAs). SLAs typically include things like availability response time for requests, assigned clients. try best exploit their by allocating them users enhances utilization lowers idle time. be done using scheduling algorithms. In proposed work" or servers maintain clients read write. Servers classified as read-and-write separately based capacity, tasks accordingly. The model uses every server's full potential, resulting an environment consumes less energy. objective Computing Resource Scheduling (GCRS) reduce consumed emitted center through resources.

Language: Английский

---

A tri-chromosome-based evolutionary algorithm for energy-efficient workflow scheduling in clouds

Swarm and Evolutionary Computation, Journal Year: 2024, Volume and Issue: 91, P. 101751 - 101751

Published: Oct. 21, 2024

Language: Английский

---