DSXchange

For your information, we make it a practice on this forum to not assist unqualified individuals in achieving positions or contracts. Not only do they rob the individuals who have taken the time to learn the skills necessary, but they lower the power scale and set an expectation of cheap services. If...

Do a "cat /.dshome" to see where the file is located. This is the "engine" directory. You do not need to recycle services to make changes effective in the odbc.ini file or the uvconfig files.

There is no driver. You must install the Oracle client software on your DataStage server. Once that is done, your administrators need to make sure that the tnsnames.ora file is current and has the appropriate connections defined. The DataStage server will use this file for connections, as well as th...

'Weird' or 'sporadic' unexplained 'what just happened' type issues usually indicate that somethings mind was blown. I see that when a high number of jobs are executing and lots of dynamic hash files are being used. This points a greasy finger straight to T30FILES. 15 dynamic hash lookups in an insta...

Sure, jobs have a command line interface accessible thru a server executable called "dsjob". You can read your manual about this for more detail. If you search this forum for "dsjob", you'll see it's been widely covered. I pasted a ksh script that shows how dsjob can be used via an enterprise schedu...

Have you considered sending output to an aggregator stage of your various metrics (row counts, etc)? Then output the aggregator data to another sequential file to have your metrics. I've done job designs where a "ready" type file is created after the job is finished writing the main file. I stream t...

Are you having any specific problems? Are you running Glance and watching cpu, disk, and memory utilization and seeing anomalous readings?

You are trolling for a lawsuit. No matter how nicely you promise, you are violating the software licensing agreement. Ascential does NOT offer a developer license for individuals such as yourself, so therefore your request cannot be fulfilled. You jeopardize this forum as a participant in facilitati...

The filename is a job parameter, and the job parameter value is fed by the job control process. The job control process takes care of determining the value by whatever means necessary. Your choices for job control are Sequencers, BATCH, or custom control using the documented APIs.

Here's some posts that cover SCDs and their implementation using DataStage and alll the nasty stuff you need to think about:

http://www.dsxchange.com/viewtopic.php?t=86234

http://www.dsxchange.com/viewtopic.php?t=85542

http://www.dsxchange.com/viewtopic.php?t=85569

http://www.dsxchange.com/viewtopic.php?t=85822

http://www.dsxchange.com/viewtopic.php?t=86187

http://www.dsxchange.com/viewtopic.php?t=84998

http://www.dsxchange.com/viewtopic.php?t=86681

http://www.dsxchange.com/viewtopic.php?t=85970

Set the constraint on your bulk load stage to "1<>1" which ensures no rows go down this path but the control file gets created when the job starts. Save some machine cycles! :D Don't evaluate 1<>1. Just use 0 (or the system variable @FALSE, which the compiler replaces with 0) as the constraint expr...

I did not see a question here. Are you asking how to design the ETL jobs to support such a task? I would start first with your modeling. You forgot in your model to include a surrogate key for ProductID, maybe ProductKey would be a good name. You also have chosen to use NULL as a startdate. This mea...

Your first approach clearly demonstrates what I am trying to tell you: DataStage is single-threaded on Server jobs. Therefore, no matter how fast Oracle can prepare and spool the data, there is a process bound by one cpu to transform it and spool its results to the loading stage. In this case, its a...

Well, DataStage Server jobs are single-threaded. This means that the SELECT query is prepared however the database does it, then a process spools the data to DataStage, which is handled in a single-threaded fashion. DS spools the data to a file, then initiates a sqlldr session to load it. Compare th...

Ken Very cool solution. How fast is it? Have you done comparisons to a multiple row lookup? I would think it is a magnitude or more faster. With preload to memory caching, you would expect it to be lightning fast, and it is. One client benchmarked a single OCI lookup using a between statement at 10...