DSXchange

Use the function to determine that the data is valid, and to determine that it matches your required pattern.

Almost certainly it was that the data were not sorted as asserted.

When you specify "Sort" on the input of an Aggregator stage you are telling DataStage that the data are already sorted, not that you require the data to be sorted.

A hashed file is a database table in the "DataStage database". It is local to the application, and employs a very fast mechanism for key-based lookups. Access it using a Hashed File stage and that very fast mechanism can be even faster, because you can cache the table in memory. Simply populate the ...

That, of course, is your call. I'm sorry you feel this way. Premium membership supports the bandwidth charges needed to keep this site running; you can link to a description of it from the home page. I guess you'll have to use your support provider, whom you pay maintenance, to provide support.

Even if they did, you'd complain that you don't like the format of the parameter file they would require you to use. My guess is that this is one of the reasons the functionality is not available "out of the box".

Welcome aboard. What do you need to achive with the lookup - merely a test for existence, or to return some columns? Are the two sources in the same Oracle instance? If so, why not perform a join (inner or left outer as appropriate) in your source stage? This will benefit from indexing on the joinin...

Auto puts:

round robin except as below

Entire on reference inputs to Lookup stage

Hash on inputs to Join and Merge stage types

This is NOT object programming. It is procedural programming. There is no place for a New declaration in DataStage BASIC. The correct syntax is: hJob = DSAttachJob("MyJobName", DSJ.ERRWARN) * Test whether attach was successful Name = DSGetJobInfo(hJob, DSJ.JOBNAME) If Name ...

... unless the granularity is too coarse!

Someone else will need to address this question; I am unfamiliar with the Oracle nomenclature. But it looks like your thoughts are on the right track.

Yay!! :applause:

"Hash" file is just plain wrong. Unfortunately, it's wrong even in some of the vendor's documentation.

Welcome aboard.
Rashmi, this forum is for questions relating to DataStage TX, not to DataStage parallel jobs. Please post your questions in an appropriate forum. Otherwise you risk getting replies about using TX maps and cards, which will serve only to confuse you.

Probably that Load/Append finishes faster, giving less opportunity for the pipe to break (or to be broken).

Most of the web clients I've encountered throw (small) XML documents at the web service. Hence my approach with the XML stages. Beware that this was a highly simplified discussion - obviously there is more detail at each step, and - as they say - the devil is in the details. A parallel job can be se...

For a start 2500 would imply GROUP.SIZE 2.

The impact can be huge - can even preclude use of dynamic hashed files!

Be very, very careful.