DSXchange

From where did you import the table definition?

When you are selecting from tables, locks ought not to be an issue; the database server will take care of this. Of course, if you specify a high transaction isolation level, then the database server may implement that using locks. When you are inserting/updating/deleting, the rules for locking are t...

What about leap years?

For 24 months wouldn't it be easier to add 2 to the year component, then roll back one day if the result is 29 Feb?

This algorithm fails for the year 2BC but otherwise is OK. You could build in a check for this if you are likely to process dates in that range.

Note that the argument to ls here is the numeric "1", not the lower case "L". The "1" option forces output to be one file per line, that is, in a single column. It's the easiest way to get a list of file names to process. You can follow it with any regular expression for filtering the list of file n...

Beware, too, that there are some errors in the Parallel Job Developer's Guide - some functions are described there that do not actually exist for the Modify stage. The preferred reference at the moment is the Orchestrate Operators Guide.

Merge can have separate reject links for each separate reference input. Join has no reject link.
Lookup has only one, irrespective of the number of reference inputs.

Of course it's possible. But there's no point.

"Am not quite aware of the hardware configuration of the mainframes and AIX box. "

Therefore I am not quite aware whether you'll see any difference in whatever you understand performance to be.

Use a job sequence with a loop including execution of your job, calculation of the next start time, and sleep.

Schedule this job sequence to start at 08:00 daily, for example.

You can view the code of ExecSH yourself - it is in the Routines branch. "Success" is driven by the exit status of the command provided via the Input Value field - what you would get with echo $? - 0 means success, a non-zero value indicates failure of some kind.

Then you may need to reset the sequence name(s) in a before-stage subroutine. Search the forum for "SDKSequences".

Use pre-SQL to execute TRUNCATE TABLE - it's much quicker.

You can not use DELETE on a file opened with OPENSEQ. You can only use DELETE on a file opened with OPEN or OPENPATH, which the documentation tells you; the file variable is assigned "in a previous OPEN statement". If you are seeking a way to delete a line from a text file, you have essentially two ...

That is a completely unfair question! What are the hardware configurations? What processing restrictions do you have on either machine? Are your DataStage jobs optimal? Will you still be sourcing data from the mainframe or is everything happening on the AIX machine? What - precisely - do you mean by...

This has been answered more than once in the past. Please search the forum.