Abstract
The tasks of data-parallel computation jobs come up with diverse and time-varying resource requirements. The dynamic nature of task requirements brings challenges on making good scheduling decisions. In this paper, we present BETS to cope with the requirement dynamics that aims at maximizing cluster resource utilization. BETS employs a task model that represents runtime task requirements, a coarse-grained task pipeline to make fully use of resources in a time-division multiplexing fashion, and fine-grained resource management to guarantee performance.
Author supplied keywords
Cite
CITATION STYLE
Li, Z., Zhang, Y., Zhao, Y., Peng, Y., & Li, D. (2016). Best effort task scheduling for data parallel jobs. In SIGCOMM 2016 - Proceedings of the 2016 ACM Conference on Special Interest Group on Data Communication (pp. 555–556). Association for Computing Machinery, Inc. https://doi.org/10.1145/2934872.2959047
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
