DSXchange

Are you sure the values are empty? Some non-printable characters don't display correctly in datastage, but are seen as values. Try either getting the length on those strings to make sure they're empty or write it to a file and vi it.

We've been using the lookup file stage instead of doing lookups from our databases and we're seeing an increase in performance (this is not using a flat file for a lookup though). You can also partition the lookup fileset if the lookup is large to keep your memory usage down. For large lookups that ...

It's like ArndW said, your job/project format default settings have the timestamp format of day then month (%dd/%mm) You can either change them to be month then day (%mm/%dd) or you can explicitly state the format of your date field in the transformer like ArndW previously posted.

If you're using the 8.0 version of Parallel you can setup a Parameter set that each of your jobs can use. That paramater set will contain the same connection information that each job can use. If you're using 7.5.x then you'll want to setup job parameters that use the same connection information (Us...

It's pretty easy to handle XML in parallel. You don't use the sequential file unless you want to stream the data into the xml stage as a full xml document record. I wouldn't recommend that. You should be able to give it a path to the xml document. You can use the list approach in a previous post to ...