Document Type
Conference Proceeding
Publication Date
12-3-2014
Journal / Book Title
2014 IEEE 7th International Conference on Cloud Computing
Abstract
Hadoop is an emerging framework for parallel big data processing. While becoming popular, Hadoop is too complex for regular users to fully understand all the system parameters and tune them appropriately. Especially when processing a batch of jobs, default Hadoop setting may cause inefficient resource utilization and unnecessarily prolong the execution time. This paper considers an extremely important setting of slot configuration which by default is fixed and static. We proposed an enhanced Hadoop system called FRESH which can derive the best slot setting, dynamically configure slots, and appropriately assign tasks to the available slots. The experimental results show that when serving a batch of MapReduce jobs, FRESH significantly improves the makespan as well as the fairness among jobs.
DOI
10.1109/CLOUD.2014.106
Montclair State University Digital Commons Citation
Wang, Jiayin; Yao, Yi; Mao, Ying; Sheng, Bo; and Mi, Ningfang, "FRESH: Fair and Efficient Slot Configuration and Scheduling for Hadoop Clusters" (2014). Department of Computer Science Faculty Scholarship and Creative Works. 296.
https://digitalcommons.montclair.edu/compusci-facpubs/296
Published Citation
J. Wang, Y. Yao, Y. Mao, B. Sheng and N. Mi, "FRESH: Fair and Efficient Slot Configuration and Scheduling for Hadoop Clusters," 2014 IEEE 7th International Conference on Cloud Computing, Anchorage, AK, USA, 2014, pp. 761-768, doi: 10.1109/CLOUD.2014.106.