Altruistic Scheduling In Multi-Resource Clusters
OSDI'16: Proceedings of the 12th USENIX conference on Operating Systems Design and Implementation(2016)
摘要
Given the well-known tradeoffs between fairness, performance, and efficiency, modern cluster schedulers often prefer instantaneous fairness as their primary objective to ensure performance isolation between users and groups. However, instantaneous, short-term convergence to fairness often does not result in noticeable long-term benefits. Instead, we propose an altruistic, long-term approach, CARBYNE, where jobs yield fractions of their allocated resources without impacting their own completion times. We show that leftover resources collected via altruisms of many jobs can then be rescheduled to further secondary goals such as application-level performance and cluster efficiency without impacting performance isolation. Deployments and large-scale simulations show that CARBYNE closely approximates the state-of-the-art solutions (e.g., DRF [27]) in terms of performance isolation, while providing 1.26 x better efficiency and 1.59 x lower average job completion time.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络