DSXchange

Try this: // First, load input date string into a date variable APT_Date tmpDate1; tmpDate1.set(in.dtfromfile, "%mm/%dd/%yyyy"); // Then, create new temp variable using Y and M pieces of first date, // and 1 for day. The class APT_Date has a constructor that we can // use: APT_Date(int yea...

One final note - the goal of multiple readers is to take advantage of multiple nodes - instead of one single stream of data, you have as many as there are readers. On the other hand, if you are using either filesets or the file pattern, you will already be taking advantage of multiple nodes - one pe...

transferRecord(); outputRecord(); You don't need these unless you are trying manually write to the output link. If you have autowrite on/true, then you don't need this. We use these function with autowrite off/false when we have a 1-to-many relationship between input and output records (ex. splitti...

Nrevezzo,

We installed the patch you recommended and now my job is working great! Thanks for the info, you saved a lot of confused head scratching.

What is your DB2 stage doing? Is it a write? Can you provide more details?

Type: String The beginning part of the name of the series of consecutively numbered columns (named like xxxyyy0 to xxxyyyn) to be combined into a vector called xxxyyy. This is taken right from Designer, the information box describing the "Column's Common Partial Name". If your input field...

Hi lshort, Look into the CDC stage. The documentation in parjdev.pdf gives additional information on CDC stage. It is used to identify new inserts, updates as well as deletes. Rich Pardon a potentially dumb question. What is the CDC Stage? I couldn't find references to it in the parjdev manual. Is ...

Hi, The stage you need is PxSurrogateKeyGenerator not column generator, column generator is a dev/debug tool. IHTH, Is there a reason why the Column Generator should not be used outside of development? Just wondering... From what I've seen, the Column Generator is far more flexible than the PxSurro...

nrevezzo wrote: Ecase 68549 is the fix. Contact Ascential support to get the patch

Thanks for the patch info. Our software guy will be working on getting that installed. We'll see if that takes care of it.

Well, unfortunately, it really is a (slow!) single node 7-AMP system: *** Query completed. 7 rows found. 2 columns returned. *** Total elapsed time was 1 second. vproc count ------ ----------- 0 1973 1 1848 2 1848 3 1847 4 1848 5 1846 6 1847 So, how did you come up with 16 for RequestedSessions and ...

The variable is APT_RDBMS_COMMIT_ROWS. We usually set it to something divisible by 4096 (why? who know). Either 8192 or 16384 are common in our processes.

You can put a query in a parameter, but you will probably run into a similar problem due to the length of the query. Off the top fo my head, I don't know what the max length of a parameter is but I'd be willing to bet that it is shorter than your query. Another possibility, one that my project uses ...

There are many options 1) You could embed the query in a view and use the same DB2 read stage, just accessing the view instead of running the actual query. 2) Put the query into a stored procedure and return a result set. Not sure which version of PX you are using, but I think hte latest (7.5) has a...

Unfortunately I already thought of that (sorry, didn't mention it in the first posting). The tables I found were <database>.terasync and <database>.ORCH_WORK_######## (where #'s appear to be a generated unique id per execution). Are those the tables you were thinking of? I have tried running both wi...

Can I add a final note? We are a banking institution as well and all our numbers are stored as decimal. For performance and Torrent conversion issues, we do not use Transforms, we use buildops. If we do calculations, we convert temporarily to float and then back to decimal. As a very simple example,...