An evolutionary method based on Lottery algorithm for independent task scheduling in cloud computing systems has been presented in this study. Task scheduling is a major issue in large-scale distributed systems that impresses on system performance. For some reasons such as heterogeneous and dynamic features in cloud environment, task scheduling is an NP-hard optimization problem. In fact, the goal is determining a processing resource from set of resources that a task needs for processing, so more jobs could be in less time. Hence task scheduling in cloud computing is very important which try to determine an efficient scheduling and source allocation. Our proposed algorithm with the base of lottery and by some goal oriented operations such as, making an optimize initial population, and also running the tasks by a special ordering considering resource load balancing and quality of service, achieves the optimize makespan and response time. It also decreases the probability of task failure rate on running based on the resource failure frequency rate, and also decreases the task starvation problem and supports the scheduling for new entered tasks in system by a dynamic method. The experimental results show that the algorithm is effective when compared with existing algorithms.
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|