A Practical Approach to NFS Response Time Monitoring

Website Maintenance Alert

Due to scheduled maintenance, the USENIX website may not be available on Monday, March 17, from 10:00 am–6:00 pm Pacific Daylight Time (UTC -7). We apologize for the inconvenience and thank you for your patience.

If you would like to register for NSDI '25, SREcon25 Americas, or PEPR '25, please complete your registration before or after this time period.

Authors: 

Gary L. Schaps & Peter Bishop, Cirrus Logic, Inc.

Abstract: 

NFS and the automounter simplify both filesystem access and filesystem management in large, distributed networks. In demanding environments, however, they require a network management strategy which ensures a consistent and acceptable level of performance.

This paper reviews available tools for measuring NFS response time including rpcspy, nfswatch, nfsstat, and WireTap. We then describe a heuristic technique we have implemented on top of basic NFS response time statistics. It is used to create a management measure of network quality based upon NFS response time, and to proactively focus on NFS performance bottlenecks.

BibTeX
@inproceedings {260227,
author = {Gary L. Schaps and Peter Bishop},
title = {A Practical Approach to {NFS} Response Time Monitoring},
booktitle = {USENIX Seventh System Administration Conference (LISA 93)},
year = {1993},
address = {Monterey, CA},
url = {https://www.usenix.org/conference/lisa-93/practical-approach-nfs-response-time-monitoring},
publisher = {USENIX Association},
month = nov
}
Download

Links

Paper: 
http://usenix.org/publications/library/proceedings/lisa93/full_papers/schaps.pdf