ACM Home Page
Please provide us with feedback. Feedback
Scale and performance in a distributed file system
Full text PdfPdf (2.38 MB)
Source ACM Transactions on Computer Systems (TOCS) archive
Volume 6 ,  Issue 1  (February 1988) table of contents
Pages: 51 - 81  
Year of Publication: 1988
ISSN:0734-2071
Authors
John H. Howard  Carnegie-Mellon Univ., Pittsburgh, PA
Michael L. Kazar  Carnegie-Mellon Univ., Pittsburgh, PA
Sherri G. Menees  Carnegie-Mellon Univ., Pittsburgh, PA
David A. Nichols  Carnegie-Mellon Univ., Pittsburgh, PA
M. Satyanarayanan  Carnegie-Mellon Univ., Pittsburgh, PA
Robert N. Sidebotham  Carnegie-Mellon Univ., Pittsburgh, PA
Michael J. West  Carnegie-Mellon Univ., Pittsburgh, PA
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 34,   Downloads (12 Months): 437,   Citation Count: 330
Additional Information:

abstract   references   cited by   index terms   review   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/35037.35059
What is a DOI?

ABSTRACT

The Andrew File System is a location-transparent distributed tile system that will eventually span more than 5000 workstations at Carnegie Mellon University. Large scale affects performance and complicates system operation. In this paper we present observations of a prototype implementation, motivate changes in the areas of cache validation, server process structure, name translation, and low-level storage representation, and quantitatively demonstrate Andrews ability to scale gracefully. We establish the importance of whole-file transfer and caching in Andrew by comparing its performance with that of Sun Microsystems NFS tile system. We also show how the aggregation of files into volumes improves the operability of the system.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
BROWNBRIDGE, D. R., MARSHALL, L. F., AND RANDELL, B. The Newcastle connection. Softw. Pract. Exper. 12 (1982), 1147-1162.
 
2
KAZAR, M.L. Synchronization and caching issues in the Andrew file system. Tech. Rep. CMU- ITC-058, Information Technology Center, Carnegie Mellon Univ., Pittsburgh, Pa., June, 1987.
3
4
 
5
RIFKIN, A. P., FORBES, M. P., HAMILTON, R. L., SABRIO, M., SHAH, S., AND YUEH, K. RFS architectural overview. In Usenix Conference Proceedings (Atlanta, Ga., Summer 1986). USENIX Association, Berkeley, Calif., pp 248-259.
6
7
 
8
SIDEBOTHAM, a.N. Volumes: The Andrew file system data structuring primitive. European Unix User Group Conference Proceedings (August 1986). Also available as Tech. Rep. CMU- ITC-053, Information Technology Center, Carnegie Mellon Univ., Pittsburgh, Pa.
 
9
SUN MICROSYSTEMS, INC. Networking on the SUN Workstation. Sun Microsystems, Mountain View, Calif., 1986.
10
 
11
TICH~, W. F., AND RUAN, Z. Towards a distributed file system. Tech. Rep. CSD-TR-480, Computer Science Dept., Purdue Univ., West Lafayette, In. 1984.
12

CITED BY  330


REVIEW

"Charles William Bash : Reviewer"

The Andrew file system described in this paper is a location-transparent distributed file system that is planned to span more than 5,000 workstations. Such a scale affects performance and complicates system operation. The authors discuss several  more...

Collaborative Colleagues:
John H. Howard: colleagues
Michael L. Kazar: colleagues
Sherri G. Menees: colleagues
David A. Nichols: colleagues
M. Satyanarayanan: colleagues
Robert N. Sidebotham: colleagues
Michael J. West: colleagues