Sprite position statement: Use distributed state for failure recovery

Brent Welch; Mary Baker; Fred Douglis; John Hartman; Mendel Rosenblum; John Ousterhout

Sprite position statement: Use distributed state for failure recovery

Brent Welch, Mary Baker, Fred Douglis, John Hartman, Mendel Rosenblum, John Ousterhout

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

Abstract

The authors advocate keeping state in main memory instead of logging state to disk so that higher performance services can be implemented. They are motivated by their distributed file system which uses stateful servers to support a high-performance distributed caching system. For reliability, a server's state is replicated in the main memory of other hosts so that the system can recover from the failure of a server. After a server reboots, its clients help it rebuild its internal state. The authors point out that as networks and processors get faster, but disks do not, relying on other hosts will be more efficient than using disks.

Original language	English (US)
Title of host publication	Proc Second Workshop Workstation Oper Sys WWOS II
Editors	Anon
Publisher	Publ by IEEE
Pages	130-133
Number of pages	4
ISBN (Print)	081862003X
State	Published - 1989
Externally published	Yes
Event	Proceedings of the Second Workshop on Workstation Operating Systems (WWOS-II) Workstation Operating Systems (WWOS-II) - Pacific Grove, CA, USA Duration: Sep 27 1989 → Sep 29 1989

Publication series

Name	Proc Second Workshop Workstation Oper Sys WWOS II

Other

Other	Proceedings of the Second Workshop on Workstation Operating Systems (WWOS-II) Workstation Operating Systems (WWOS-II)
City	Pacific Grove, CA, USA
Period	9/27/89 → 9/29/89

ASJC Scopus subject areas

General Engineering

Cite this

Welch, B, Baker, M, Douglis, F, Hartman, J, Rosenblum, M & Ousterhout, J 1989, Sprite position statement: Use distributed state for failure recovery. in Anon (ed.), Proc Second Workshop Workstation Oper Sys WWOS II. Proc Second Workshop Workstation Oper Sys WWOS II, Publ by IEEE, pp. 130-133, Proceedings of the Second Workshop on Workstation Operating Systems (WWOS-II) Workstation Operating Systems (WWOS-II), Pacific Grove, CA, USA, 9/27/89.

@inproceedings{aaa1890a55a544199d7f5b8ce856f301,

title = "Sprite position statement: Use distributed state for failure recovery",

abstract = "The authors advocate keeping state in main memory instead of logging state to disk so that higher performance services can be implemented. They are motivated by their distributed file system which uses stateful servers to support a high-performance distributed caching system. For reliability, a server's state is replicated in the main memory of other hosts so that the system can recover from the failure of a server. After a server reboots, its clients help it rebuild its internal state. The authors point out that as networks and processors get faster, but disks do not, relying on other hosts will be more efficient than using disks.",

author = "Brent Welch and Mary Baker and Fred Douglis and John Hartman and Mendel Rosenblum and John Ousterhout",

year = "1989",

language = "English (US)",

isbn = "081862003X",

series = "Proc Second Workshop Workstation Oper Sys WWOS II",

publisher = "Publ by IEEE",

pages = "130--133",

editor = "Anon",

booktitle = "Proc Second Workshop Workstation Oper Sys WWOS II",

note = "Proceedings of the Second Workshop on Workstation Operating Systems (WWOS-II) Workstation Operating Systems (WWOS-II) ; Conference date: 27-09-1989 Through 29-09-1989",

}

TY - GEN

T1 - Sprite position statement

T2 - Proceedings of the Second Workshop on Workstation Operating Systems (WWOS-II) Workstation Operating Systems (WWOS-II)

AU - Welch, Brent

AU - Baker, Mary

AU - Douglis, Fred

AU - Hartman, John

AU - Rosenblum, Mendel

AU - Ousterhout, John

PY - 1989

Y1 - 1989

N2 - The authors advocate keeping state in main memory instead of logging state to disk so that higher performance services can be implemented. They are motivated by their distributed file system which uses stateful servers to support a high-performance distributed caching system. For reliability, a server's state is replicated in the main memory of other hosts so that the system can recover from the failure of a server. After a server reboots, its clients help it rebuild its internal state. The authors point out that as networks and processors get faster, but disks do not, relying on other hosts will be more efficient than using disks.

AB - The authors advocate keeping state in main memory instead of logging state to disk so that higher performance services can be implemented. They are motivated by their distributed file system which uses stateful servers to support a high-performance distributed caching system. For reliability, a server's state is replicated in the main memory of other hosts so that the system can recover from the failure of a server. After a server reboots, its clients help it rebuild its internal state. The authors point out that as networks and processors get faster, but disks do not, relying on other hosts will be more efficient than using disks.

UR - http://www.scopus.com/inward/record.url?scp=0024941928&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024941928&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0024941928

SN - 081862003X

T3 - Proc Second Workshop Workstation Oper Sys WWOS II

SP - 130

EP - 133

BT - Proc Second Workshop Workstation Oper Sys WWOS II

A2 - Anon, null

PB - Publ by IEEE

Y2 - 27 September 1989 through 29 September 1989

ER -

Sprite position statement: Use distributed state for failure recovery

Abstract

Publication series

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this