5 Tips For Better DataStage Design #18

1. Use Database Connector Stages wherever possible
    - Older “Enterprise” stages must not be used
    - ODBC stages must not be used when Connectors are available 
    - Transformer & Basic Transformer Stage Usage
    - Transformer usage should be minimized (they are CPU intensive)
        * If other stages such as Copy can accomplish the same task use those instead
    - Basic Transformers must be removed (they don't work in Grid)

2. Lookup vs. Join 

    - If reference data is small  and source data is large Lookup should be used
    - If reference data is large and source data is small Join should be used
    - If both reference data and source data are small Join should be used

3. Non-sequence jobs should not have more than 15 stages.

4. “BULK” Mode should be used in Teradata Loads
    - Specify TENACY and SLEEP parameters if needed to prevent failed loads
    - Specify “Yes” to Parallel synchronization
    - NOTE: Teradata info may be refined based on feedback from EDW and Teradata teams

5. Oracle Connectors must read/write in parallel
    - Execution mode on Advanced tab must be set to “parallel”
    NOTE:  Oracle info may be refined based on FPD Oracle testing in Grid

6. DB2 Connectors must read/write in parallel   
    - Enable partitioned reads must be set to yes



Like the below page to get update  
https://www.facebook.com/datastage4you
https://twitter.com/datagenx
https://plus.google.com/+AtulSingh0/posts
https://datagenx.slack.com/messages/datascience/