DSXchange

Following will explain what is reverse lookup Dimension Table : DIM_DATE DIM_Date_Surrogate_Key Date_Natural_Key 1 20040721 2 20040722 3 20040723 Business Data(Which is the source data) Rec_No Sales_Date Sale_Value 100 20040721 400.00 200 20040721 600.00 300 20040722 700.00 Based on your Business Da...

When you use the update existing/insert new option from the drop down list datastage doesn't do anything to your database. It sends the records to the table like usual update sql and insert sql. Based on your database and the way you created the table the new records will go accordingly.

Thanks
Siva

The question was not clear enough to answer. With DB2 plug-in you can do a full replace to the existing table. There are more options available when you use db2 like insert/update/replace etc. To know more about the options read the db2 bulk load manual.

Thanks
Siva

Search for the topic slowly changing dimension you will find many threads with that discussion in this forum.

Thanks
Siva

Another advantage is that you get separate logs in your director for each instance. This is very useful to check what is the result for each invocation id's.

Thanks
Siva

Assign a parameter in your job (FileDateName) and before running the job pass the current date to the parameter and start your job. In the sequential stage put the parameter after the file("d:\datastage\output\file_#FileDateName#.csv

Thanks
Siva

Goto the command line and type uv to enter universe from there type DS.TOOLS which will bring you the menu. Type QUIT to come out of universe. Alternatively you can use the Datastage Administer to run the command. Go to program folder of Datastage and inside click Datastage ADministrator from there ...

Send the insert and update to two different text file and use them in the same job to do insert and update. This will be the work around if you want that to happen in the same job.

Cheers
Rasi

This happens basically becos of the locks in database. The best practise is to split the two stream (insert/update) separate and then do the inserts or updates. If you want this to happen in the same job then send the target to text file and then use the same text file and do your inserts and update...

Instead of doing the update and Insert in the same job. Try outputting the target to a sequential file and then do the update following by insert job. This should rectify your problem.



Thanks
Rasi

Hi Vipin

Below is the contact details of Ascential

http://www.ascential.com/company/local.html

you can call and talk the authorised person to get the price information.

Cheers
Siva

You can create your own Naming standard to your business and put it in a document so that everyone who develops the job follows the standard. To start from you can have the project name(devt/test/prod), Job Category (Based on your groups), Job Names, Parameters (working directory/hash dir/database n...

This can be done using Pivot Stage in Datastage. Use the Student Id as the key columns and the marks as derived field as(Sub1,Sub2,Sub3mSub4). This is will create four records for you. There is Pivot Stage PDF document which explains about this stage with example.

Cheers
Siva

Hi Suresh, In Datastage its upto in what way you want to process. Main dataset you call it as main stream and the lookup is called as reference. If you have your main from database or from any source the lookup can also be any source(hash or database). All your main dataset passes from source to tar...

Yes it is possible to do in DataStage. It is just a matter of passing the parameter and using the correct conv functions to do. According to your parameter the conversion will change.

Cheers
Rasi