AlloX: compute allocation in hybrid clusters
AlloX: compute allocation in hybrid clusters
复制标题
DOI：
10.1145/3342195.3387547
复制
发表时间：
2020-04
期刊：
影响因子：
--
通讯作者：
T. Le;Xiao Sun;Mosharaf Chowdhury;Zhenhua Liu
中科院分区：
文献类型：
--
作者：
T. Le;Xiao Sun;Mosharaf Chowdhury;Zhenhua Liu
现代深度学习框架支持多种硬件，包括CPU、GPU和其他加速器，以执行计算。在本文中，我们研究了如何在这些可互换的资源上调度作业-每个资源具有不同的计算速度-以优化性能，同时在共享集群中提供用户之间的公平性。我们从理论和经验上证明，现有的解决方案及其直接的修改在存在可互换资源的情况下表现不佳，这促使了allX的设计和实现。其核心是将调度问题转化为最小代价的二部匹配问题，并提供随时间的动态公平分配。我们从理论上证明了它在理想的离线环境下的最优性，并通过与Kubernetes的结合，经验地证明了它在在线场景中工作得很好。在一个小规模的CPU-GPU混合集群上的评估和大规模模拟表明，allX可以显著减少平均作业完成时间(当系统负载较高时，最高可减少95%)，同时提供公平性和防止饥饿。
Modern deep learning frameworks support a variety of hardware, including CPU, GPU, and other accelerators, to perform computation. In this paper, we study how to schedule jobs over such interchangeable resources - each with a different rate of computation - to optimize performance while providing fairness among users in a shared cluster. We demonstrate theoretically and empirically that existing solutions and their straightforward modifications perform poorly in the presence of interchangeable resources, which motivates the design and implementation of AlloX. At its core, AlloX transforms the scheduling problem into a min-cost bipartite matching problem and provides dynamic fair allocation over time. We theoretically prove its optimality in an ideal, offline setting and show empirically that it works well in the online scenario by incorporating with Kubernetes. Evaluations on a small-scale CPU-GPU hybrid cluster and large-scale simulations highlight that AlloX can reduce the average job completion time significantly (by up to 95% when the system load is high) while providing fairness and preventing starvation.