Cloud infrastructures are suitable environments for processing large scientific workflows. Nowadays, new challenges are emerging in the field of optimizing workflows such that it can meet user’s service quality requirements. The key to workflow optimization is the scheduling of workflow tasks, which is a famous NP-hard problem. Although several methods have been proposed based on the genetic algorithm for task scheduling in clouds, our proposed method is more efficient than other proposed methods due to the use of new genetic operators as well as modified genetic operators and the use of load balancing routine. Moreover, a solution obtained from a heuristic used as one of the initial population chromosomes and an efficient routine also used for generating the rest of the primary population chromosomes. An adaptive fitness function is used that takes into account both cost and makespan. The algorithm introduced in this paper utilizes a load balancing routine to maximize resources’ efficiency at execution time. The performance of the proposed algorithm is evaluated by comparing the results with state of the art algorithms of this field, and the results indicate that the proposed algorithm has remarkable superiority in comparison to other algorithms and performs task scheduling with the least makespan and cost.

云基础架构是处理大型科学工作流程的合适环境。如今，在优化工作流以使其能够满足用户的服务质量要求方面出现了新的挑战。工作流程优化的关键是工作流程任务的调度，这是一个著名的NP难题。尽管已经提出了几种基于遗传算法的云任务调度方法，但是由于使用了新的遗传算子，改进的遗传算子和负载均衡例程，因此我们提出的方法比其他提出的方法效率更高。此外，从启发式方法获得的解决方案可用作初始种群染色体之一，而高效例程也用于生成其余的主要种群染色体。使用了一种自适应适应度函数，该函数同时考虑了成本和制造时间。本文介绍的算法利用负载平衡例程来最大化执行时的资源效率。通过将结果与该领域的最新算法进行比较，评估了该算法的性能，结果表明，与其他算法相比，该算法具有显着的优势，并且以最小的制造时间和成本执行任务调度。