Earticle

초록

영어

Hadoop uses a reliable, efficient and scalable way to process data. It provides a good solution for dealing with big data. The task scheduler is the core component of Hadoop, and it is responsible for the managing and allocating the cluster resources. Therefore, scheduling algorithm directly affects the overall performance of Hadoop platform and utilization of cluster resource. Based on this, the improved accelerate particle swarm algorithm (IAPSO) is introduced to the cloud environment, and to solve the cloud task scheduling problem in this article. When we use particle swarm algorithm for task scheduling, the tasks are considered as particles, the resource pool is seen as the search space, and the process of finding the optimal solution is considered as a process of task scheduling. If all the sub tasks find the appropriate resources, then stop the iteration and allocate sub asks to the resource nodes. Finally, we simulate the experiment by using CloudSim software. When a single type of task is committed, our algorithm and the other three algorithms can also be used to complete the task scheduling process, and our algorithm is more efficient. But in practice, the cloud computing environment is facing multiuser, and the types of tasks are also varied. With the increase in the number of tasks, the advantage of the other three algorithms decreases gradually, but algorithm in this paper has been exhibited higher efficiency. In addition, with the increase of the number of nodes, task completed time of the algorithm in this paper is significantly less than the other three algorithms, and it has a steady downward trend. Therefore, IAPSO algorithm which is proposed in this paper is applied to solve task scheduling problem in the cloud environment, and it can effectively improve the efficiency of task scheduling.

목차

Abstract
 1. Introduction
 2. Task Scheduling in Hadoop Architecture
  2.1. Hadoop Distributed File System
  2.2. Mapreduce Model
 3. The Standard PSO Algorithm
  3.1. The Architecture of PSO
  3.2. The Basic Formula of PSO and its Improved Form
 4. The Cloud Task Scheduling Model Based on Hadoop is Constructed by Using PSO Algorithm
 5. The Simulation and Result Analysis
  5.1 Task Type Setting
  5.2. The Simulation
 6. Conclusion
 Acknowledgments
 References

키워드

저자

• Zhulin Li [ College of information science and engineering, Northeastern University, Shenyang, Liaoning, China, Modern educational technology center, Shenyang Agricultural University, Shenyang, Liaoning, China ] Corresponding author
• Cuirong Wang [ College of computer and communication engineering, Northeastern University at Qinhuangdao, Qinhuangdao, Hebei, China ]
• Haiyan Lv [ College of forestry, Shenyang Agricultural University, Shenyang, Liaoning, China ]
• Tongyu Xu [ College of information and electrical engineering, Shenyang Agricultural University, Shenyang, Liaoning, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Grid and Distributed Computing

• 간기

  격월간

• pISSN

  2005-4262

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605