DSXchange

You have DataStage, so you do have DataStage shell.

Isolate the date and time components. Convert them into internal format, convert the days into seconds, and perform the arithmetic. The result is in seconds - it can be subsequently converted with simple arithmetic using standard arithmetic.

The Remove Duplicates stage (which requires sorted input) gives you the ability to specify preservation of the First or Last record in each group. The Sort stage (with "unique" set) does not.

Did the job report use of block overflow? If not you're within reasonable sizes.

Welcome aboard.

RESIZE command must be issued in a DataStage shell (dssh), not in a DOS shell (cmd).

Do not change the uvconfig entry.

If you want to create a 64-bit hashed file, you can't do it with the hashed file stage, but you can do it through the ExecTCL before-job subroutine.

This is not useful information from the point of view of being a competent DataStage developer or even administrator. When a job is compiled, what happens depends upon the job type. It's different for all four job types. Indeed, the very notion of "back end" may be different - mainframe jo...

ODBC protocols refer to "data sources" rather than "servers". Of course you can have multiple DSNs for the same database, but it doesn't gain anything except, perhaps, greater ease/understanding of management.

The process table is part of the operating system, not of DataStage. The process ID is effectively the row identifier in this table (although memory is managed better than that; rows are removed from memory when a process terminates).

Node, in DataStage parallel execution parlance, is one subdivision of processing and data partitioning. It has no relationship at all with hardware configuration. There may be many processes running on each node and, as Craig noted, it's up to the operating system on that node's machine (its fastnam...

Have you tried using the Oracle Connector stage (new in version 8.1)?

Yes, but Oracle only gives a single stream from the result set.

Have you set the Partition Table property?

There's not really enough information - or, it depends.

For example, do you have "always running" or "start on demand" jobs? Are they multi-instance or not?

That'll do it nearly every time!

Of course it is, but what the stage variable - or even the system variable - will have is the total number of rows so far . So you would need an Aggregator to trap just the final value. And then you have the issue of how to get both values into the same row in the table without causing locking probl...

That's how the Oracle Enterprise stage works, unless the table is a partitioned table and you've programmed that knowledge into the stage.