DHPC Adelaide

DHPC Technical Report DHPC-030

Querying and Auxiliary Data in the DISCWorld

J.A.Mathew and K.A.Hawick

Archived: 22 January 1998

Published in Proc. of the 5th IDEA Workshop, Fremantle, February 1998.

Abstract

An integral part of the DISCworld environment is access to the storage and manipulation of the auxiliary data that is associated with each of the large data objects that is stored in a repository. For efficient job execution, it is necessary to minimise the movement and processing of relatively large files: this can be accommodated by pre-identifying files of interest by querying the auxiliary data assocaited with each file. A natural storage means of such auxiliary data would be in a RDBMS. However the interfaces to most RDMSs is proprietary and requires vendor and platform specific software to be installed at both the client and server. The Java JDBC package provides a mechanism for universal access to the data stored in a database. We propose, and have built a prototype implementation for, a high-level Java-based database layer that allows for storage and querying of auxiliary data. It also provides mechanisms by which the schema of the database can be dynamically altered, and uses Boolean Syntax Query form in preference to the SQL that is native to the database. Our implementation uses Informix Universal Server as the underlying database and we suggest how the Object-Relational extensions that this database provides can be utilised in constructing a storage environment for auxiliary data. We also discuss how graphical Java client applications can be constructed so users can browse auxiliary data without prior knowledge of the schema.

HTML version

PDF version

PostScript version (gzip compressed)


[ DHPC Adelaide | DHPC Bangor | Contacts | People | Projects | Reports ]

webmaster@dhpc.adelaide.edu.au