DSXchange

Up the minimum modulus from 1. What you're probably seeing is the resize by doubling effect degrade performance. Checkout this post as well:
http://www.dsxchange.com/viewtopic.php?t=85364

As much as I enjoy this thread, we're discussing last years model. The Server product is not the future, it's the Parallel edition. The flexibility and ease of use of the Server product will allow you to do anything. Every custom solution has to be done with an eye towards open metadata and sharing....

I have a rules/metadata based data cleansing module/bolt-on for DataStage to enforce L0 data quality with some cross-column verification and enforcement. However, the question is transformation, not cleansing. That being said, T is mostly about enforcing business rules, from my experience. The clean...

Search the forum for mkdbfile command. You can manually created the hash file with the 64bit keyword so that you don't have to mess around with all of these afterwards commands.

The introduction of a rules based engine for transformation removes the T from the ETL formula. In my opinion, the T is the strength of DataStage. I suggest that your company probably bought the wrong tool if all they need DataStage to handle is the E and the L.

Unsorted data requires the aggregator to work much harder, and there are limitations as to how much unsorted data it can aggregate. If you have 1 million rows that group to 1 million rows, the aggregator will have performance issues. If you have 1 million rows that group to 10 thousand rows, the agg...

Load only and UPSERT are vastly different actions in the database. UPSERT involves DML, where Load uses sqlldr. So, in a nutshell, you're comparing DML loading versus sqlldr, and probably sqlldr DIRECT path.

Where's nag?

Anywhere you need to format a value such as a derivation, key expression, stage variable, or constraint:

Search the forum for DEFFUN and you'll see examples

should be

Let us know if your problem is still around.

I bet the T30FILES value is too low in the uvconfig file. You probably will need to up this if you're using a lot of dynamic hash files. Remember, each job uses a couple of dynamic hash files just by their nature (logs, status, config, etc), in addition to any your design uses.

Write Cache will not overwrite a hash record so the first record in is the final record only on duplicate keys. Going back in time a little, but this came up in conversation this morning... This statement concerned me, so I wrote a little test case to see if it was really the case that overwrites d...