DSXchange

Have you considered the Merge stage to read the two files, then running the result through a sort?

Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518

As above plus:
The InterCall manual has the most comprehensive list of error codes for client/server connections.
The BCI manual is useful for those wishing to use BCI functions from DataStage routines.


Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518

All UniVerse manuals can be found on the IBM web site.
http://www-3.ibm.com/software/data/u2/universe
Possibly the best manuals for DataStage are those for UniVerse release 9.6, which are at
http://www-3.ibm.com/software/data/u2/p ... index.html


Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518

telnet is different on UNIX and Windows. On UNIX, telnet is an operating system service, so you could use dsjob directly after connecting. On Windows, telnet is a DataStage service (dstelnet), so you would need to escape to an operating shell to execute dsjob. For example: DOS /C "dsjob ..." As you ...

Would really need to see both jobs to make any accurate diagnosis. Why not do separate exports of each, and compare the DSX files using diff? This might reveal the awful truth, whatever it is. The behaviour is redolent of UniVerse behaviour when trying to process a non-existent Select List; all reco...

Once I was involved with exposing DB2/400 tables on DB2 on an RS6000. One of the things we found was that, because every file on the AS400 is a table, it was vital to specify a filter on table names before connecting, otherwise the connection awaited the gathering of every single file name on the AS...

For this approach to work, you will need to have used a pattern to the names of your hashed files, and they will have to have been created with CREATE.FILE in the local account. Otherwise you will need to adjust the code to use DELETE.FILE, DROP TABLE or rm (or del) as appropriate. If there is no pa...

You can only retrieve a job's user status while you have a handle to the job (that is, the job is attached). You may need to re-attach the job to achieve this. The DSGetJobInfo() function is how you retrieve the job's user status value. Check out the help or the manual for full details. Then take a ...

IF, most of the I/O statements and a number of other statements can be in "single-line format" or "multi-line format". The rule that governs it can be stated as follows: If the clause-introducing keyword (for example ON ERROR, LOCKED, THEN or ELSE) is the final word on the line in which it appears, ...

For some reason there are two threads with this topic. I responded to the other one, with why the Hashed File is not appropriate for a BETWEEN join. Admin, can you merge these when populating the archive?

Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518

Actually, since parallel jobs are working with data sets (which are in memory as much as possible) there are about four different strategies that you might use for lookup, for example a MERGE.

Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518

The whole point of a hashed file is to perform an equi-join on the primary key. This value is passed through a hashing algorithm to generate the physical address of where the record in question is stored, without the need to consult an index nor to perform a table scan. This is an extremely fast str...

Are you asking the question as a newbie to DataStage, or someone who has been using server jobs in the past? The number 1 difference is that you must do a bit more planning, related to resource allocation and data partitioning. From these plans you construct your configuration files (you may have di...

Can anybody accurately differentiate agreement/disagreement weights and match weights? In INTEGRITY SuperMATCH output, once records are blocked, you see some form of composite weight beside EVERY record. Is this the match weight and, if so, why is there one beside the XA records? Surely an XA record...

Enrol in the DataStage Best Practices (Server) class (DS306). There you will learn good ways to implement all three types of slowly changing dimensions, as well as at least three ways to generate surrogate keys.
I believe that DS306 is available via "e-Learning" as well.