The Zebra Striped Network File System

John H Hartman, John K. Ousterhout

Research output: Contribution to journalArticle

90 Citations (Scopus)

Abstract

Zebra is a network file system that increases throughput by striping the file data across multiple servers. Rather than striping each file separately, Zebra forms all the new data from each client into a single stream, which it then stripes using an approach similar to a log-structured file system. Thm provides high performance for writes of small files as well as for reads and writes of large files. Zebra also writes parity information in each stripe in the style of RAID disk arrays; this increases storage costs slightly, but allows the system to continue operation while a single storage server is unavailable. A prototype implementation of Zebra, built in the Sprite operating system, provides 4-5 times the throughput of the standard Sprite file system or NFS for large files and a 15-300%. improvement for writing small files.

Original languageEnglish (US)
Pages (from-to)274-310
Number of pages37
JournalACM Transactions on Computer Systems (TOCS)
Volume13
Issue number3
DOIs
StatePublished - Jan 8 1995

Fingerprint

Servers
Throughput
Costs

Keywords

  • Log-based striping
  • log-structured file system
  • parity computation
  • RAID

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

The Zebra Striped Network File System. / Hartman, John H; Ousterhout, John K.

In: ACM Transactions on Computer Systems (TOCS), Vol. 13, No. 3, 08.01.1995, p. 274-310.

Research output: Contribution to journalArticle

Hartman, John H ; Ousterhout, John K. / The Zebra Striped Network File System. In: ACM Transactions on Computer Systems (TOCS). 1995 ; Vol. 13, No. 3. pp. 274-310.
@article{d4130da4af39415f883feaf6e4d765da,
title = "The Zebra Striped Network File System",
abstract = "Zebra is a network file system that increases throughput by striping the file data across multiple servers. Rather than striping each file separately, Zebra forms all the new data from each client into a single stream, which it then stripes using an approach similar to a log-structured file system. Thm provides high performance for writes of small files as well as for reads and writes of large files. Zebra also writes parity information in each stripe in the style of RAID disk arrays; this increases storage costs slightly, but allows the system to continue operation while a single storage server is unavailable. A prototype implementation of Zebra, built in the Sprite operating system, provides 4-5 times the throughput of the standard Sprite file system or NFS for large files and a 15-300{\%}. improvement for writing small files.",
keywords = "Log-based striping, log-structured file system, parity computation, RAID",
author = "Hartman, {John H} and Ousterhout, {John K.}",
year = "1995",
month = "1",
day = "8",
doi = "10.1145/210126.210131",
language = "English (US)",
volume = "13",
pages = "274--310",
journal = "ACM Transactions on Computer Systems",
issn = "0734-2071",
publisher = "Association for Computing Machinery (ACM)",
number = "3",

}

TY - JOUR

T1 - The Zebra Striped Network File System

AU - Hartman, John H

AU - Ousterhout, John K.

PY - 1995/1/8

Y1 - 1995/1/8

N2 - Zebra is a network file system that increases throughput by striping the file data across multiple servers. Rather than striping each file separately, Zebra forms all the new data from each client into a single stream, which it then stripes using an approach similar to a log-structured file system. Thm provides high performance for writes of small files as well as for reads and writes of large files. Zebra also writes parity information in each stripe in the style of RAID disk arrays; this increases storage costs slightly, but allows the system to continue operation while a single storage server is unavailable. A prototype implementation of Zebra, built in the Sprite operating system, provides 4-5 times the throughput of the standard Sprite file system or NFS for large files and a 15-300%. improvement for writing small files.

AB - Zebra is a network file system that increases throughput by striping the file data across multiple servers. Rather than striping each file separately, Zebra forms all the new data from each client into a single stream, which it then stripes using an approach similar to a log-structured file system. Thm provides high performance for writes of small files as well as for reads and writes of large files. Zebra also writes parity information in each stripe in the style of RAID disk arrays; this increases storage costs slightly, but allows the system to continue operation while a single storage server is unavailable. A prototype implementation of Zebra, built in the Sprite operating system, provides 4-5 times the throughput of the standard Sprite file system or NFS for large files and a 15-300%. improvement for writing small files.

KW - Log-based striping

KW - log-structured file system

KW - parity computation

KW - RAID

UR - http://www.scopus.com/inward/record.url?scp=84976704061&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84976704061&partnerID=8YFLogxK

U2 - 10.1145/210126.210131

DO - 10.1145/210126.210131

M3 - Article

VL - 13

SP - 274

EP - 310

JO - ACM Transactions on Computer Systems

JF - ACM Transactions on Computer Systems

SN - 0734-2071

IS - 3

ER -