Datastage repository behind the scenes.

Marley777 · Post by **Marley777** » Fri Dec 10, 2004 1:00 pm

Hello everyone

Thanks for your time.

Behind the scenes is the Datastage repository a file structure or is it DB2 tables? I'm not talking about the client side. I need to know the behind the scenes technical stuff.

kcbland · Post by **kcbland** » Fri Dec 10, 2004 1:29 pm

It's a code branch of IBMs Universe product. Check it out.

ray.wurlod · Post by **ray.wurlod** » Fri Dec 10, 2004 1:58 pm

You may "need to know" (why?), but Ascential elects not to publish this information and reserves the right to change it at any time.

In quick summary it's not file-based, it's database-based. It's definitely NOT DB2; Ascential explicitly markets itself as "database agnostic". To that end, currently, it uses a proprietary database called DataStage Engine that is based on IBM's UniVerse RDBMS.

Originally, and through to version 5.2, DataStage used UniVerse as its actual engine.

You can research this history on this forum. Search for VMARK, Ardent and UniVerse.

Currently there are two main sets of tables. Tables with names beginning "DS_" contain design-time information, while tables with names beginning "RT_" contain run-time information.

Expect major changes to the Repository structure in the next release, and further changes beyond that.