Reducing file system latency using a predictive approach

James Griffioen; Randy Appleton

Reducing file system latency using a predictive approach

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

211 Scopus citations

Abstract

Despite impressive advances in file system throughput resulting from technologies such as high-bandwidth networks and disk arrays, file system latency has not improved and in many cases has become worse. Consequently, file system I/O remains one of the major bottlenecks to operating system performance [10]. This paper investigates an automated predictive approach towards reducing file latency. Automatic Prefetching uses past file accesses to predict future file system requests. The objective is to provide data in advance of the request for the data, effectively masking access latencies. We have designed and implement a system to measure the performance benefits of automatic prefetching. Our current results, obtained from a trace-driven simulation, show that prefetching results in as much as a 280% improvement over LRU especially for smaller caches. Alternatively, prefetching can reduce cache size by up to 50%.

Original language	English
Title of host publication	USENIX Summer 1994 Technical Conference
ISBN (Electronic)	1880446626, 9781880446621
State	Published - 1994
Event	USENIX Summer 1994 Technical Conference - Boston, United States Duration: Jun 6 1994 → Jun 10 1994

Publication series

Name	USENIX Summer 1994 Technical Conference

Conference

Conference	USENIX Summer 1994 Technical Conference
Country/Territory	United States
City	Boston
Period	6/6/94 → 6/10/94

Bibliographical note

Funding Information:
This work was supported in part by NSF grant number CCR-9309176

Funding Information:
This work was supported in part by NSF grant number CCR-9309176 We would like to thank the reviewers for their helpful comments and suggestions. We would also like to thank Mary Baker for reviewing an early draft of the paper and providing valuable feedback. Finally we would like to thank the DCS users for submitting to being traced.

Publisher Copyright:
© 1994 USENIX Association. All rights reserved.

ASJC Scopus subject areas

Computer Science (all)

Cite this

@inproceedings{d95b634516894baa8703c32ad0dabea3,

title = "Reducing file system latency using a predictive approach",

abstract = "Despite impressive advances in file system throughput resulting from technologies such as high-bandwidth networks and disk arrays, file system latency has not improved and in many cases has become worse. Consequently, file system I/O remains one of the major bottlenecks to operating system performance [10]. This paper investigates an automated predictive approach towards reducing file latency. Automatic Prefetching uses past file accesses to predict future file system requests. The objective is to provide data in advance of the request for the data, effectively masking access latencies. We have designed and implement a system to measure the performance benefits of automatic prefetching. Our current results, obtained from a trace-driven simulation, show that prefetching results in as much as a 280% improvement over LRU especially for smaller caches. Alternatively, prefetching can reduce cache size by up to 50%.",

author = "James Griffioen and Randy Appleton",

note = "Funding Information: This work was supported in part by NSF grant number CCR-9309176 Funding Information: This work was supported in part by NSF grant number CCR-9309176 We would like to thank the reviewers for their helpful comments and suggestions. We would also like to thank Mary Baker for reviewing an early draft of the paper and providing valuable feedback. Finally we would like to thank the DCS users for submitting to being traced. Publisher Copyright: {\textcopyright} 1994 USENIX Association. All rights reserved.; USENIX Summer 1994 Technical Conference ; Conference date: 06-06-1994 Through 10-06-1994",

year = "1994",

language = "English",

series = "USENIX Summer 1994 Technical Conference",

booktitle = "USENIX Summer 1994 Technical Conference",

}

TY - GEN

T1 - Reducing file system latency using a predictive approach

AU - Griffioen, James

AU - Appleton, Randy

N1 - Funding Information: This work was supported in part by NSF grant number CCR-9309176 Funding Information: This work was supported in part by NSF grant number CCR-9309176 We would like to thank the reviewers for their helpful comments and suggestions. We would also like to thank Mary Baker for reviewing an early draft of the paper and providing valuable feedback. Finally we would like to thank the DCS users for submitting to being traced. Publisher Copyright: © 1994 USENIX Association. All rights reserved.

PY - 1994

Y1 - 1994

N2 - Despite impressive advances in file system throughput resulting from technologies such as high-bandwidth networks and disk arrays, file system latency has not improved and in many cases has become worse. Consequently, file system I/O remains one of the major bottlenecks to operating system performance [10]. This paper investigates an automated predictive approach towards reducing file latency. Automatic Prefetching uses past file accesses to predict future file system requests. The objective is to provide data in advance of the request for the data, effectively masking access latencies. We have designed and implement a system to measure the performance benefits of automatic prefetching. Our current results, obtained from a trace-driven simulation, show that prefetching results in as much as a 280% improvement over LRU especially for smaller caches. Alternatively, prefetching can reduce cache size by up to 50%.

AB - Despite impressive advances in file system throughput resulting from technologies such as high-bandwidth networks and disk arrays, file system latency has not improved and in many cases has become worse. Consequently, file system I/O remains one of the major bottlenecks to operating system performance [10]. This paper investigates an automated predictive approach towards reducing file latency. Automatic Prefetching uses past file accesses to predict future file system requests. The objective is to provide data in advance of the request for the data, effectively masking access latencies. We have designed and implement a system to measure the performance benefits of automatic prefetching. Our current results, obtained from a trace-driven simulation, show that prefetching results in as much as a 280% improvement over LRU especially for smaller caches. Alternatively, prefetching can reduce cache size by up to 50%.

UR - http://www.scopus.com/inward/record.url?scp=85084161825&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85084161825&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85084161825

T3 - USENIX Summer 1994 Technical Conference

BT - USENIX Summer 1994 Technical Conference

T2 - USENIX Summer 1994 Technical Conference

Y2 - 6 June 1994 through 10 June 1994

ER -

Reducing file system latency using a predictive approach

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this