DSXchange

The result set of the query should stream into DataStage as fast as DB2 can deliver. The problem is somewhere else in your job design. Prove this with a job designed as DB2 -----> Peek with your join query in the DB2 stage. This job will show you that the join is not the issue, if what you say about...

You marked your job type as "parallel" but posted your question in the server forum. Which is the correct job type?

Something like the following may suffice. Using Equate allows the output file delimiter characters to be somewhat negotiable. The OpenTextFile (or perhaps OpenSequentialFile) routine can be found by searching the forum. FUNCTION Cartesian(Arg1, Arg2) DEFFUN OpenTextFile(FilePath, OpenMod...

Have your DB2 DBA show you that you are introducing self-locking processes. Then design your job in a way that this does not occur. Take note of the advice given by Arnd and Craig.

Don't re-open the file for each iteration of the outer loop. You can reset the file pointer using the SEEK statement. This piece of code: OpenSeq File1 To FileVar Then End means that your routine attempts to keep going whether or not the file is opened. This is not good practice! There is nowhere in...

You can even store all your parameters in a flat file and pass this file to -param argument. The format of the parameter file should be: paramater1=value parameter2=value parameter3=value .. .. And your command would become dsjob -run -param parameterfile.txt project job Is this documented somewher...

Again, probably easiest done with SQL rather than with DataStage. Form a join of the table to itself, based on id being the same and element being different. SELECT t1.ID, t1.ELEMENT, t2.ELEMENT FROM tablename t1, tablename t2 WHERE t1.ID = t2.ID AND t1.ELEMENT <> t2.ELEMENT; Does that fulfil your r...

As Arnd said, it's primarily about data volume. However, the parallel engine is the core functionality going forward, so if you want all of the benefits of all of the product suite, then you might lean towards parallel. That's not to say that server jobs are going away any time soon; they're not. Ho...

Welcome aboard. :D This sounds suspiciously like an interview question. A short answer is that one uses a value to make its decision, the other can use a expression that is somewhat like an SQL WHERE clause. Performance is not the issue; it's functionality. You choose the right stage for the job. Re...

Those five parameters ought to be straightforward. Read man ftp on your UNIX machine to get some feel for how FTP works. Server Name: name (or IP address) of the Windows machine - it must be configured to accept inbound FTP on port number 21 User: Windows login ID PWD: Windows password Remote Path: ...

Welcome aboard. :D

To follow Sanjeev's advice you need a UNIX-based ODBC driver for Excel.

If you do not have one, you will probably need to save the worksheet in question in delimited format, and process the resultant text file using a Sequential File stage.

It's your choice as to how to get them out of a table, presumably into script (shell) variables.

I'd prefer an approach using a job sequence, but you specified "from the operating system".

Assuming each developer logs in with a separate user ID, LIST.READU alone will identifier the owner of the lock. This may also be executed at the operating system prompt, so you can filter the output using grep. For example DSHOME=`cat /.dshome` export DSHOME . $DSHOME/dsenv $DSHOME/bin/list_readu |...

You can implement as described in the remainder of my post.

To view the full content of premium posts, sign up for premium membership. It's less than $1 per week, and the revenue is used to fund this site's continued existence.

I wonder whether a stream containing three left outer joins would work for this case?