Towards distributed machine learning in shared clusters: A dynamically-partitioned approach

• Main Authors: SUN, Peng, WEN, Yonggang, TA, Nguyen Binh Duong, YAN, Shengen
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2017
• Subjects: Artificial Intelligence and Robotics; Software Engineering
• Online Access: https://ink.library.smu.edu.sg/sis_research/4766; https://ink.library.smu.edu.sg/context/sis_research/article/5769/viewcontent/1704.06738.pdf
• Institution: Singapore Management University

Many cluster management systems (CMSs) have been proposed to share a single cluster with multiple distributed computing systems. However, none of the existing approaches can handle distributed machine learning (ML) workloads given the following criteria: high resource utilization, fair resource allocation and low sharing overhead. To solve this problem, we propose a new CMS named Dorm, incorporating a dynamicallypartitioned cluster management mechanism and an utilizationfairness optimizer. Specifically, Dorm uses the container-based virtualization technique to partition a cluster, runs one application per partition, and can dynamically resize each partition at application runtime for resource efficiency and fairness. Each application directly launches its tasks on the assigned partition without petitioning for resources frequently, so Dorm imposes flat sharing overhead. Extensive performance evaluations showed that Dorm could simultaneously increase the resource utilization by a factor of up to 2.32, reduce the fairness loss by a factor of up to 1.52, and speed up popular distributed ML applications by a factor of up to 2.72, compared to existing approaches. Dorm’s sharing overhead is less than 5% in most cases. Index Terms—Cluster Resource Management, Distributed Machine Learning, Fairness