DSXchange

The Aggregator stage calculates its results as dfloat, as a technique to avoid arithmetic overflow. But there's another property in the Aggregations folder, called Default to Decimal Output, that defaults (if I remember correctly) to 8,2. This is applied if the data type of the output is Decimal or ...

Do you have the appropriate SAP plug-in stage(s) installed?

Do you have existing projects? How many?
Has anything been moved since DataStage was installed on the server?

Server or parallel job? YOu have posted in the server job forum but designated your job type as parallel. Are there any warnings logged when you run the job? How have you set up the BAPI stage? That is, what property settings have you used? You might also need to investigate why the export paramete...

Can you combine the work of T3, T4 and T5 into one Transformer stage, to try to reduce the impact of many processes?

If not, can you increase the buffer sizes and timeout values?
Perhaps double each of them.

Are you updating the same table from which you are selecting?

This is well covered in the manuals. Adjacent active stages in general are combined into the same process. IPC stages force process boundaries to occur, as do intermediate passive stages. Link Partitioner and Collector stages do too, in addition to which they create a process of their own. Passive s...

Start over.
Document (not in pseudo code, but in English or any other spoken language) what you have to do.
This will make the design requirements easier for you, and for us, to understand.

Welcome aboard. :D Please delete duplicate post before anyone responds to it. Etiquette Note: We don't do urgent here. Your note arrived on the server at 3:46am where I was at the time. If you want urgent help, sign up for premium service from your support provider, and learn the true cost of "urgen...

Welcome aboard. :D Lose the "all Java" requirement and DataStage is an ideal tool for you. You can still use Java for the business logic, but in DataStage parallel jobs the "language" used to shift data around is built using C++ (which you don't have to write). You can initiate DataStage jobs from a...

The key performance indicator in ETL is elapsed time. If you can get the data in - or out, as the case may be - within the allowed time window, with a safety margin, then your ETL is performing satisfactorily. Fast is far less important than correct. One stream doing insert only and another stream d...

Failed to enumerate groups is not a major issue.

The original problem occurred in DSD.INTopen. Do you have two adjacent active stage types in your design? Do you have inter-process row buffering enabled?

Welcome aboard. :D The key performance indicator in ETL is elapsed time. If you can get the data in - or out, as the case may be - within the allowed time window, with a safety margin, then your ETL is performing satisfactorily. The strongest mechanism for tuning that performance, therefore, is expe...

Welcome aboard. :D The key performance indicator in ETL is elapsed time. If you can get the data in - or out, as the case may be - within the allowed time window, with a safety margin, then your ETL is performing satisfactorily. Obviously there are going to be obstacles when transferring data to dif...