DSXchange

So you are talking about a server job with the DB2 API stage right? Have you tried the same with a parallel job and the enterprise DB2 stage? It creates reject codes in a different way by propogating the SQL Code so it may be more reliable, I have found it handles larger commit sizes more effectivel...

Jim, you should post that to the FAQ forum as the solution for a vertical pivot requirement.

My tests of 2000 lookup rows or less with a dataset or sequential file as the lookup source did not create a lookuptable, my tests of large lookup volumes did create lookuptables. The smaller lookups are fitting in RAM memory making them much faster and efficient. This is why the manuals advise usin...

There are a couple people coordinating it in Australia and they have already emailed and phoned me a couple times, but I expect them to trail the US by a couple weeks in terms of getting software and getting it to people. You should ring your nearest Ascential office and try to find out who the loca...

It is an interesting question, what are lookup tables? How are the native dataset and lookup filesets treated by the lookup stage? I put a long answer together in my latest blog. There are some benefits from using the lookup fileset over the other types.

You posted in the server forum but you set the job category to parallel. Which is it? You mention DB2 partitioning is set to null but that you are using the API stage and not the enterprise stage.

You cannot create something out of nothing in a Modify stage. It cannot add columns, it can only copy columns or convert columns. Your specification needs to include an existing column name. If you want to generate a new column use the column generator stage or the transformer stage.

You should keep retrieving your surrogate key start value back at the sequence job level via a routine. It can be difficult getting the maximum key value used from within a parallel job as you have multiple instances of transformers and surrogate key generators in action and no easy way to get the l...

Doing it in a job will only work if you have a known number of output files and you can define an output link for each one. If you are talking about a dynamic number of output files then you should stick with a looping routine. If you are using a parallel job be aware that unless you force it to ret...

It might be faster with the vertical pivot, but with the horizontal pivot (even in parallel), awk is still 20 times faster. I have a case open with Ascential on the terrible performance of this stage. But this is a different issue. horizontal vs. vertical. Or, maybe it isn't. Since the stage is cal...

You need to perform a DB2 remote server configuration. Ring Ascential/IBM support and ask for the latest DB2 Remote Configuration Guide or get the old version I uploaded to Ascential devnet. You will need to add the client option to your DB2 stage which will let you define the remote user login and ...

Be very careful with CHAR and VARCHAR combinations and trimming as sometimes your before values get implicitely modified and no longer match your after values. Try removing columns from the value list one at a time via explicit value settings until you find which field is causing the problem.

You also have something call MetaArchitect that lets you write your own metabrokers. You already have metabrokers for the other Ascential products and for common BI tools such as BO and Cognos and for the common modelling tools such as ErWin.

With parallel processing DataStage should be faster then an awk script. If you have a static vertical pivot, where you know how many rows belong to each key, then it is easy to do in a parallel job. If you have a dynamic number of rows for each key it becomes quite difficult. In the example link pro...

Your DataStage job will have a problem if you try to run any type of derivation on a field that passes a null value, such as a comparison or calculation. The fields that are passed straight through without transformation should be okay. You will have to post the error message for more help.