DSXchange

It depends on a number of factors, including the number of AMPs available in Teradata. Work with your DBA (and, if you have one, a DataStage practitioner experienced with Teradata).

You can get some benefit too from the IBM Knowledge Center, or the Connectivity Guide for Teradata Databases manual.

Direct the "unused" links into some kind of a sink, such as a Copy stage.

You must connect four links in the job to four links in the Shared Container.

Welcome aboard. How did you "convert to ASCII"? Did you try "read as EBCDIC" in the Sequential File stage?

The Where Used analysis would be the easier route.

If you do not have the table definition, export the project and search the export file.

Or use a utility like "files in a project" from Chuck Smith

Good call. That's why I asked the question. If you had elected to use a remote database, I would have advised strongly to use a dedicated, very high speed network segment.

Determine whether record-level or page-level deadlocks are occurring, with the help of your DBA. If it's record-level, you can change your job design so that the same key value cannot be processed on more than one partition.

What's on node LKF-PISRVENG01 ?

Tell us how fast the network is between where you propose to install the engines and where the databases are. That will be the main consideration.

You could export the project then search the export file for parametersetname.parametername

If you're running from the clients (Designer or Director), you can change your engine credentials using the web console for Information Server (on the Administration tab.)

You will, of course, need the dsadm password.

Use a second Aggregator to find those maxima.

Please tidy up your ASCII art using the Preview facility.

Then tell us whether the inputs to the Change Capture stage are identically sorted and partitioned.

Sort by key then by CRDR. Use an Aggregator stage to count groups. Don't forget to partition by key.

Initialize stage variables to 0. Detect change in key, possibly using LastRowInGroup() function. Accumulate while not last row in group; reset once last row in group has been processed. Use extra stage variables to "remember".

1. Yes.
2. No - this used to be the case, but has not been true since version 8.1.1.