FILE SYSTEM CACHING IN LARGE POINT-TO-POINT NETWORKS

P B Austin; K A Murray; A J Wellings

FILE SYSTEM CACHING IN LARGE POINT-TO-POINT NETWORKS

P B Austin, K A Murray, A J Wellings

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

In this paper, we examine the design of a file system for a general-purpose scalable parallel computing engine, based on an extensible mesh of processors. Although many distributed file systems have already been developed, none have properly addressed the characteristics that are special to these computing engines, namely dynamic load-balancing and point-to-point networks. Client caching is identified as a key area where current systems fall short. We propose two new caching schemes that employ an extra level of caching in addition to that used by current client caching schemes. We present some results from a simulation study of the two new caching schemes, which demonstrate that the schemes outperform traditional client caching schemes when used in a general-purpose processor mesh-based, parallel computing engine. Since one of these caching schemes is quite complex, we have developed a formal specification in the Z language.

Original language	English
Pages (from-to)	65-80
Number of pages	16
Journal	Software Engineering Journal
Volume	7
Issue number	1
Publication status	Published - Jan 1992

Keywords

PERFORMANCE
DESIGN

Cite this

@article{be06a2c3a208411eb5fc0f582985a425,

title = "FILE SYSTEM CACHING IN LARGE POINT-TO-POINT NETWORKS",

abstract = "In this paper, we examine the design of a file system for a general-purpose scalable parallel computing engine, based on an extensible mesh of processors. Although many distributed file systems have already been developed, none have properly addressed the characteristics that are special to these computing engines, namely dynamic load-balancing and point-to-point networks. Client caching is identified as a key area where current systems fall short. We propose two new caching schemes that employ an extra level of caching in addition to that used by current client caching schemes. We present some results from a simulation study of the two new caching schemes, which demonstrate that the schemes outperform traditional client caching schemes when used in a general-purpose processor mesh-based, parallel computing engine. Since one of these caching schemes is quite complex, we have developed a formal specification in the Z language.",

keywords = "PERFORMANCE, DESIGN",

author = "Austin, {P B} and Murray, {K A} and Wellings, {A J}",

year = "1992",

month = jan,

language = "English",

volume = "7",

pages = "65--80",

journal = "Software Engineering Journal",

issn = "2053-910X",

publisher = "Institution of Engineering & Technology",

number = "1",

}

TY - JOUR

T1 - FILE SYSTEM CACHING IN LARGE POINT-TO-POINT NETWORKS

AU - Austin, P B

AU - Murray, K A

AU - Wellings, A J

PY - 1992/1

Y1 - 1992/1

N2 - In this paper, we examine the design of a file system for a general-purpose scalable parallel computing engine, based on an extensible mesh of processors. Although many distributed file systems have already been developed, none have properly addressed the characteristics that are special to these computing engines, namely dynamic load-balancing and point-to-point networks. Client caching is identified as a key area where current systems fall short. We propose two new caching schemes that employ an extra level of caching in addition to that used by current client caching schemes. We present some results from a simulation study of the two new caching schemes, which demonstrate that the schemes outperform traditional client caching schemes when used in a general-purpose processor mesh-based, parallel computing engine. Since one of these caching schemes is quite complex, we have developed a formal specification in the Z language.

AB - In this paper, we examine the design of a file system for a general-purpose scalable parallel computing engine, based on an extensible mesh of processors. Although many distributed file systems have already been developed, none have properly addressed the characteristics that are special to these computing engines, namely dynamic load-balancing and point-to-point networks. Client caching is identified as a key area where current systems fall short. We propose two new caching schemes that employ an extra level of caching in addition to that used by current client caching schemes. We present some results from a simulation study of the two new caching schemes, which demonstrate that the schemes outperform traditional client caching schemes when used in a general-purpose processor mesh-based, parallel computing engine. Since one of these caching schemes is quite complex, we have developed a formal specification in the Z language.

KW - PERFORMANCE

KW - DESIGN

M3 - Article

SN - 2053-910X

VL - 7

SP - 65

EP - 80

JO - Software Engineering Journal

JF - Software Engineering Journal

IS - 1

ER -