DSXchange

Looks like a straightforward pivot to me, after you've generated the column heading "values".

Job parameter?

Sample stage.

The space is the sign character. The only way you can lose it is to convert the data type to Char (or VarChar) and strip the white space.

Check out Metadex from CompactBI. It does more than what you ask but, if you've got a lot of table definitions and maps to ingest, it may save you some work.

Just use the DataSet Management Tool (or orchadmin command) to copy the Data Sets.

Specify, in the Sequential File stage, that the file contains a byte order mark.

Batch jobs differ from regular jobs for execution only in that their names contain "::". You can load batch jobs into Job activities just like other kinds of job.

The question can be answered by interrogating IBM's website, so what benefit is there in posting it here?

Not without writing your own parallel routines, one of which would be a "rules dispatcher".

Why not use a Sequential File stage? These have a property that allows them to read from or write to named pipes.

Actually the first thing that came to my mind was the question "are you processing the same data in each run?"

Lots of gotchas, mainly relating to host names, IP addresses, and the resolutions thereof. These are recorded in the metadata repository, so things break if they're not in the correct location.

Can you produce just the bitmap using dsdesign.exe?

The location for Parameter Sets associated with jobs is a subdirectory in the project. This is non-optional, as it allows them to be backed up as "project directories" using istool. The location for named parameter sets is a subdurectory in the DSEngine directory. This, too, is not negotia...