Search found 42189 matches
- Tue Jan 02, 2007 10:25 am
- Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
- Topic: Getting number of rows for Stages or Links
- Replies: 3
- Views: 1917
- Sun Dec 31, 2006 8:33 am
- Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
- Topic: Convert function in a MF job
- Replies: 6
- Views: 2656
- Fri Dec 29, 2006 8:22 pm
- Forum: General
- Topic: Happy New Year, 2007.
- Replies: 16
- Views: 6135
- Fri Dec 29, 2006 8:14 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Is there a way to invoke a Stored Procedure from a ctl Job ?
- Replies: 2
- Views: 757
- Fri Dec 29, 2006 4:16 pm
- Forum: General
- Topic: Happy New Year, 2007.
- Replies: 16
- Views: 6135
- Fri Dec 29, 2006 4:09 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Not able to Oper a Job
- Replies: 6
- Views: 2053
- Fri Dec 29, 2006 4:01 pm
- Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
- Topic: multiple rows from a row based on values in a field
- Replies: 11
- Views: 2361
- Fri Dec 29, 2006 3:58 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: Not able to Oper a Job
- Replies: 6
- Views: 2053
- Fri Dec 29, 2006 3:52 pm
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: inserting very very large no records
- Replies: 6
- Views: 1453
Meaning you mark the job as 'Allow Multiple Instance' as noted. Since each Invocation ID you use must be unique across all running instances, why not leverage it to also control what happens in your job? In general, it could be a filename or table name or something you use to constrain a source quer...
- Fri Dec 29, 2006 11:08 am
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: inserting very very large no records
- Replies: 6
- Views: 1453
I'd suggest you look into a multi-instance job where the InvocationID macro is used to control the partition being loaded. Then rather than running one large job with all the associated issues with restarting that you'd have, run multiple instances instead. Errors within a single partition can be de...
- Fri Dec 29, 2006 10:37 am
- Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
- Topic: multiple rows from a row based on values in a field
- Replies: 11
- Views: 2361
Yes, you need to figure out the logic needed to do that. Or are you just going to wait for someone to hand it to you on a silver platter? Perhaps one approach would be to treat each record as comma delimited and then parse the first field into two pieces around the tilde. Do you know what the maximu...
- Fri Dec 29, 2006 10:34 am
- Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
- Topic: many records to one
- Replies: 7
- Views: 2020
In other words, you'll need to build a single record from the multiple incoming records yourself via concatenation in stage variables. As noted, you can search the forum for many discussions on the subject of a 'vertical pivot' which is what that would be called when you do rows to columns. Columns ...