DSXchange

XMETA has a minimum version of Oracle 10g anyway, so that would be OK.

There is no way you can force there to be the same number of rows on each of four nodes if the total number of rows is not a multiple of 4.

Modify stage can not, as a general rule, perform operations that take two inputs. Therefore concatenation is not an operation that a Modify stage can perform. Transformer is the correct stage for the task. You could create your own (Build) stage, you could use an External Filter stage, but Transform...

select CustID, Count(*) from table group by CustID having Count(*) > 1; In DataStage split your stream of rows into two using a Copy stage, and count the rows for each CustID group in an Aggregator on one of those streams before joining the two streams (using a Join stage) on CustID. Filter only th...

Read the file using a single VarChar column. Your file is NOT fixed-length format, which would require every record to contain the same number of characters. Specify none for the field delimiter. Determine the record type in a Filter stage or in Transformer stage constraints and parse the separate r...

Is this question about InfoSphere Transformation Extender (TX) or about DataStage parallel jobs?

Please post in the Enterprise Edition forum if it is not about TX - the forum in which you did post is for that product only. The product used to be called DataStage TX but is no longer thus called.

ArndW wrote:Why not use an inner join in a join stage or a lookup stage (choose either link as a master) and have both Col2's go to the output?

A full outer join would be needed to achieve all of the output specified. Other than that quibble I, too, would recommend using a Join stage.

The "fork join" model suggested by Sainath seems to me to be the way to go.

A Hashed File is the mechanism by which database tables in the UniVerse database (and therefore in the DataStage Engine database) are implemented. You can do anything with a hashed file that you can do with any other database table stage, and a couple more things as well. It's "hashed", no...

He did. Check whether ODBC drivers are installed.

The UniVerse connection already exists, and is called localuv.

... and, within the ELSE path of the OPENSEQ statement, you can use a CREATE statement to force the file to exist.

Welcome aboard.

Is this an Information Analyzer project or a project for some other product within IBM Information Server?

Have you specified a length property for every field?

It IS trouble-free. But there's no performance gain - indeed there appears to be a slight performance diminution, probably caused by needing to interface with the Information Server services and the remote repository.