Affiliation:
1. NetApp, Sunnyvale, CA
2. Johns Hopkins Univeristy
Abstract
We develop a holistic framework for adaptively scheduling asynchronous requests in distributed file systems. The system is holistic in that it manages all resources, including network bandwidth, server I/O, server CPU, and client and server memory utilization. It accelerates, defers, or cancels asynchronous requests in order to improve application-perceived performance directly. We employ congestion pricing via online auctions to coordinate the use of system resources by the file system clients so that they can detect shortages and adapt their resource usage. We implement our modifications in the Congestion-Aware Network File System (CA-NFS), an extension to the ubiquitous network file system (NFS). Our experimental result shows that CA-NFS results in a 20% improvement in execution times when compared with NFS for a variety of workloads.
Funder
Division of Computing and Communication Foundations
Division of Information and Intelligent Systems
Publisher
Association for Computing Machinery (ACM)
Subject
Hardware and Architecture
Reference43 articles.
1. Amir Y. Awerbuch B. Danilov C. and Stanton J. 2005. A cost-benefit flow control for reliable multicast and unicast in overlay networks. Trans. Netw. 10.1109/TNET.2005.857121 Amir Y. Awerbuch B. Danilov C. and Stanton J. 2005. A cost-benefit flow control for reliable multicast and unicast in overlay networks. Trans. Netw. 10.1109/TNET.2005.857121
2. Anderson E. Spence S. Swaminathan R. Kallahalla M. and Wang Q. 2005. Quickly finding near-optimal storage designs. Trans. Comput. Syst. 10.1145/1113574.1113575 Anderson E. Spence S. Swaminathan R. Kallahalla M. and Wang Q. 2005. Quickly finding near-optimal storage designs. Trans. Comput. Syst. 10.1145/1113574.1113575
3. Measurements of a distributed file system
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献