DSXchange

Why?

What kind of "binary"? After all, every decimal number is already in a binary form within DataStage.

Do you need some form of "packed decimal", or perhaps binary coded decimal representation?

I guess the easiest way is to write via a Complex Flat File stage.

Research the rtype options for conversions such as DecimalToDecimal().

Jump in your time machine and go back to the run before the abort. This is the only way you can find out for sure.

Easy-peasy.

Export the job into an XML file.

I am a server job guru (!) and wish to clarify as follows. If two Transformer stages - or any kind of active stages - are directly connected by a link then, by default, they will run in the same process. Enabling inter-process row buffering, whether at the job level or by placing an IPC stage betwee...

The database has been created with no logging, as you note. I am not aware of any mechanism for suppressing this alert message. If there is, it will be an Informix environment variable.

Not in server jobs, I suspect.

Server jobs - you need to create separate jobs.

There are alternatives in parallel jobs, but this is not that forum.

At one point the hashed file (note - it's not "hash" file) was not completely deleted.

All DELETEs will be logged - that is, you will always "get a log". DB2 does not have the option of disabling transaction logging. Auto-commit (rows/transaction = 1) means that the number of pending rows will be minimized, so you don't risk blowing your undo tablespace. Based on far too lit...

Moderator: please move to server forum

Any answer you get in this (parallel job) forum will bewilder you further.

Have you tried limiting the Netezza stage to a single-node node pool?

Yep. Multiple ReadV is a waste of I/O. Do a single Read, then use Field() or Extract() to pull out the individual multi-valued fields. Then use Field() or Extract() or angle-bracket notation to access the individual values. All Collections in DataStage are stored in the local repository as either as...