Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
skip to main content
article

Boosting I/O performance of internet servers with user-level custom file systems

Published: 01 September 2001 Publication History

Abstract

Previous studies have shown that disk I/O times are one of the major performance bottlenecks of Internet servers such as proxy cache servers. Most conventional file systems do not work well for such systems because of their very high overheads. Although Special-purpose operating systems may achieve high performance, it is very difficult and expensive to design and maintain. They also have very poor portability. In this paper we propose to built user-space, customized file systems for Internet servers so as to achieve high-performance, low-implementation-cost and good portability at the same time. To provide an example of such systems, we presented a novel scheme called WPSFS that can drastically improve I/O performance of proxy servers and other applications. WPSFS is an application-level software component of a proxy server which manages data on a raw disk or disk partition. Since the entire system runs in the user space, it is easy and inexpensive to implement. It also has good portability and maintainability. With efficient in-memory meta-data data structures and a novel file system called Page-structured file system(PFS), WPSFS achieves 9-20 times better I/O performance than the state-of-the-art SQUID server running on a Unix Fast File System, and 4-10 times better than the improved SQUIDML.

References

[1]
J. Almeida and P. Cao, "Measuring proxy performance with the wisconsin proxy benchmark," Tech. Rep. 1373, Computer Science Department, University of Wisconsin-Madison, April 1998.
[2]
C. Maltzahn, K. J. Richardson, and D. Grunwald, "Reducing the disk I/O of web proxy server caches," in Proceedings of the 1999 USENIX Annual Technical Conference (USENIX-99), (Berkeley, CA), pp. 225-238, USENIX Association, June 6-11 1999.
[3]
D. Wessels, "Squid frequently asked questions," 2001. http://www.squidcache.org/Doc/FAQ/FAQ.html.toc13.
[4]
M. Rosenblum and J. K. Ousterhout, "The design and implementation of a log-structured file system," ACM Transactions on Computer Systems, vol. 10, pp. 26-52, Feb. 1992.
[5]
T. Nightingale, Y. Hu, and Q. Yang, "The design and implementation of DCD device driver for UNIX," in Proceedings of the 1999 USENIX Technical Conference, (Monterey, California), pp. 295-308, Jan. 1999.
[6]
G. R. Ganger, M. K. McKusick, C. A. N. Soules, and Y. N. Patt, "Soft updates: a solution to the metadata update problem in file systems," ACM Transactions on Computer Systems, vol. 18, pp. 127-153, May 2000.
[7]
M. F. Kaashoek, D. R. Engler, G. R. G. anger, and D. A. Wallach, "Server operating systems," in 1996 SIGOPS European Workshop, Sept. 1996.
[8]
P. Cao and S. Irani, "Cost-aware WWW proxy caching algorithms," in Proceedings of the USENIX Symposium on Internet Technologies and Systems (ITS-97), (Berkeley), pp. 193-206, USENIX Association, Dec. 8-11 1997.
[9]
J. Wang, R. Min, Z. Wu, and Y. Hu, "WPSFS - a user-space, high performance, custom file system for web proxy servers," Tech. Rep. TR258/04/01ECECS, Department of Ele & Computer Engineering and Computer Science, University of Cincinnati, Feb. 2001.
[10]
B. Jenkins, "A new hash functions for hash table lookup," Dr. Dobb's Journal, Sept 1997.
[11]
G. R. Ganger and Y. N. Patt, "Using system-level models to evaluate I/O subsystem designs," IEEE Transactions on Computers, vol. 47, pp. 667-678, June 1998.
[12]
D. Wessels, "NLANR CACHE README," April 1998. ftp://ircache.nlanr.net/Traces/README.
[13]
IETF. http://www.ietf.org, 1997. The HTTP 1.1 Protocol-Draft.
[14]
K. Kant and P. Mohapatra, "Scalable Internet servers: issues and challenges," in Performance Evaluation Review, 2000.
[15]
E. P. Markatos, M. G. H. Katevenis, D. Pnevmatikatos, and M. Flouris, "Secondary storage management for web proxies," in Proceedings of the 2nd USENIX Symposium on Internet Technologies and Systems (USITS-99), (Berkeley, CA), pp. 93-104, USENIX Association, Oct. 11-14 1999.
[16]
A. Iyengar, S. Jin, and J. Challenger, "Efficient algorithms for persistent storage allocation," in Proceedings of the 18th of IEEE Symposium on Mass Storage Systems, (San Diego, California), April 2001.

Cited By

View all

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM SIGMETRICS Performance Evaluation Review
ACM SIGMETRICS Performance Evaluation Review  Volume 29, Issue 2
September 2001
40 pages
ISSN:0163-5999
DOI:10.1145/572317
Issue’s Table of Contents

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 01 September 2001
Published in SIGMETRICS Volume 29, Issue 2

Check for updates

Qualifiers

  • Article

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)1
  • Downloads (Last 6 weeks)0
Reflects downloads up to 16 Oct 2024

Other Metrics

Citations

Cited By

View all
  • (2023)A Review of Data Mining, Big Data Analytics and Machine Learning ApproachesJournal of Computing and Natural Science10.53759/181X/JCNS202303016(169-181)Online publication date: 5-Oct-2023
  • (2004)Reducing disk I/O times using anticipatory movements of the disk headJournal of Systems Architecture: the EUROMICRO Journal10.1016/S1383-7621(03)00097-350:1(17-33)Online publication date: 1-Jan-2004
  • (2002)Web-conscious storage management for web proxiesIEEE/ACM Transactions on Networking10.1109/TNET.2002.80483610:6(735-748)Online publication date: 1-Dec-2002
  • (2002)UCFS-A Novel User-Space, High Performance, Customized File System for Web Proxy ServersIEEE Transactions on Computers10.1109/TC.2002.103262551:9(1056-1073)Online publication date: 1-Sep-2002

View Options

Get Access

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Media

Figures

Other

Tables

Share

Share

Share this Publication link

Share on social media