DSXchange

In Australia the present ratio of server to parallel sites is about 9:1. In south and east Asia (excluding Japan and companies doing offshore work for US clients) the ratio seems even higher. This is, as Vincent noted, gradually changing, but server will be around for quite some time yet.

No.

You could script it (sqlplus) and have DataStage execute the script, but there is nothing within DataStage.

In a word, parallelism

What happens when you try it?

Does the Change Capture stage even permit nullable fields in the comparison key column(s)? Hint: review the Nulls Position property in the Change Keys group of properties.

Because there is a comma-delimited list in a single column, it is necessary first to split these into separate columns. You can do so in a Transformer stage using the Field() function.

Don't kill the UNIX processes with kill -9

Berate anyone else who does so.

What the manual is trying to suggest (and getting it just slightly wrong enough so it doesn't work) is that you can - if you wish - define the sub-record structures for the first and second columns. Even if you do this properly (in a schema definition) you must use unique column names in the separat...

It is not a DataStage problem. It is a problem with the way you are using DataStage, and the size of the transaction log. The latter can definitely be addressed by the DBA.

Linux (must be Red Hat AS 3.0)

DataStage server components for Linux (the Windows version will not install on Linux). Visit the IBM web site to view availability matrix.

You do not need to install Multi Client Manager.

Install DataStage clients on a Windows machine.

That's a different question. The OP's source was Oracle - you don't get any rejects from a SELECT statement. With a Sequential File stage you can have a rejects output link which captures, in raw format, any row that does not satisfy the metadata of the primary output link. Without a rejects output ...

The Hawk release includes some quite impressive estimation tools.

It also raises the interesting question how to create a new copy of a Data Set using the orchadmin utility, if you can't use copy unless the target already exists. So it was worth noting this bug, so that users don't get frustrated when it doesn't work.

Are you sure about the name of the environment variable? I can not find this one mentioned in the Parallel Job Advanced Developer's Guide (version 7.5x2). Do you mean APT_NO_PART_INSERTION ? Many of the environment variables involve a double negative. That is, when they are activated they disable so...

It allows more than one version of the client to be installed on the same PC. It ought not to affect installing the server.