DSXchange

What do you mean by "incorrect"? Do you get correct results if you disable parallel execution (as a test)? Have you tried explicitly partitioning on FILE_ID using a key-based algorithm? Are both inputs sorted on FILE_ID?

In your case, the last 1 record, and the tail stage must run in sequential mode so that it only extracts one "last" record.

With all those words, I can't see how it would be easier. You could pre-process INPUT to determine start and end dates.

DataStage Home Activity Discussion Groups Contests Lounge Members Ask a Question DataStage| The DataStage group is for the discussion of technical issues that arise during the installation or daily administration of IBM InfoSphere DataStage software. Home | Invite Peers | More Data Warehouse Groups...

It's not really a DataStage task, but perhaps you could perform a full outer join using a Merge stage. This will not yield differences, but you didn't ask for that.

Why not just generate all dates and perform a lookup against your INPUT?

A more flexible approach (for example per-node) is available. Will post details later (probably on the weekend). Tres busy at the moment.

Drag second link from stage, right click on second link and choose "convert to reject". Pretty much everything else happens automatically, including sql code appearing on the reject link.

If you're using SMP architecture Entire partitioning is essentially free - one copy of the Data Set is emplaced in shared memory and used by all nodes.

It's not made visible.

Have you identically trimmed both the source column value and the reference key column value?

Add an extra line terminator character to the end of each CTR record.

In server jobs the browser uses the same technique as row limiting in the Transformer stage - it's just not made visible. I think "they" assumed that no-one would want to waste their time looking at 10000 rows through this interface, which is not a particularly efficient one (a trade-off f...

All three sorts (Sort stage, input link sort and inserted sort) cause a tsort operator to be included in the score. All three are governed by environment variables such as APT_TSORT_STRESS_BLOCKSIZE, obviously, because all three use the tsort operator.