Browse Prior Art Database

Turnkey Solution to Automatically Publish Data from Any Relational Database

IP.com Disclosure Number: IPCOM000029349D
Original Publication Date: 2004-Jun-25
Included in the Prior Art Database: 2004-Jun-25
Document File: 2 page(s) / 54K

Publishing Venue

IBM

Abstract

Disclosed is a Turnkey Solution to Automatically Publish Data from Any Relational Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

Turnkey Solution to Automatically Publish Data from Any Relational Database

In today's environment, when data is shared across many different systems, the data source owner has to package the data and send it to different locations via FTP (File Transfer Protocol) or publish/subscribe (pub/sub) engine. Traditionally, this method requires the data source owner to create an application to detect changes (including inserts and deletes) and package the information to send the subscribers (typically implemented as an unique point-to-point interface between two systems). Another example involves populating a data warehouse which requires data from many data sources. Again, this requires each data source owner to create a new application or change the existing application to detect changes and send the information to the warehouse (thus multiplying the typical point-to-point interfaces, and usually with duplicated logic/effort). Traditionally, these efforts require the involvement and coordination of many development teams.

A turnkey solution is needed that can automatically detect changes from the data sources and publish these changes to the subscribers without requiring application development and redundant logic/effort.

The turnkey solution can automatically detect and publish changed data (including inserts and deletes) to subscribers. Our solution that implements this invention is based on *DB2, **Java, XML (Extensible Markup Language) and *MQSeries publish/subscribe (pub/sub) technologies, but the concept can be implemented using any publish/subscribe engine and relational databases.

The turnkey solution does not involve traditional application development to accomplish the publication of changed data. Instead, it involves simple and straightforward specification of meta-information that is then used to generate JavaBeans, XML documents, and the configuration for the publish/subscribe engine. Changed data (including inserts and deletes) from a relational database is published as XML documents with provision for subscribers to receive the XML documents as JavaBeans.

In order to detect changes from data source, the solution depends on the existence of a last updated timestamp attribute in the data source. The solution can track the last updated timestamp retrieved from each data source and retrieve only records with newer timestamp. For record deletion, the solution requires the data source to track this condition and store the information in a table (only need to store the key information that the subscribers know about)...