DSXchange

Hi All,

Is it possible to do an ImpactAnalysis on Columns,
without using MetaStage tool.

I tried using "SearchDSX" job from DeveloperNet, looks like its not working.

Any thoughts?

Thanks in Advance.

Poorna,

What specific information are you looking for to do your inpact analysis?

For example, are you trying to determine what the inpact of changing the name of a column in one of your source/target tables might be to your code?

There is a tool called Reporting Assistant that I used with some success in version 6.0. You have to explicitly install this tool from the DS CD's. It appears that the tool is no longer supported, but you may still get some value from it.

Do a search through the forums for more info.

HTH

Rob W.

The Usage Analysis tool, when run on a table definition, is as close as you'll get to an impact analysis. However, proper functioning of this hinges upon proper management of metadata in jobs - never breaking the link between the job and the table definition(s) loaded into it.

ray.wurlod wrote:The Usage Analysis tool, when run on a table definition, is as close as you'll get to an impact analysis. However, proper functioning of this hinges upon proper management of metadata in jobs - never breaking the link between the job and the table definition(s) loaded into it.

Thanks Ray & Rob.


I am looking at Export file(XML format & DSX format).
Is there a way we can extract, required information from those files?

Thanks in Advance.

It remains the case that you can not determine impact unless the table (or file) name remains associated with the job. That is, for example, you can not use a job parameter for the table (or file) name; otherwise an impact analysis can only reveal that "this job affects a table (or file) whose name is not determined until run time".

My List all files and tables used by jobs in a dsx file tool can help you identify impacts at a file and table level. Since it works with a dsx file, it does not require that you followed best practices with your metadata repository.

You will find it on the DataStage Tools page of www.anotheritco.com.

Nothing, not even Chuck's utility, can save you if you use job parameters for table names. You must - necessarily - learn only that "some table was used in this job".

chucksmith wrote:My List all files and tables used by jobs in a dsx file tool can help you identify impacts at a file and table level. Since it works with a dsx file, it does not require that you followed best practices with your metadata repository.

You will find it on the DataStage Tools page of www.anotheritco.com.

Thanks Chucksmith.

I downloaded - "Search a dsx file for a string" from www.anotheritco.com,and ran the job .

I didn't get any results from that job.