#
|
Category
|
Things to keep in mind
|
1
|
General Recommendations
|
· Reduce the number of transformations
· Calculate once, use many times – For e.g. Use Variable Ports
· Only connect what’s being used
· Facilitate reuse – Use reusable transformations, mapplets
· Filter records early-on and set driving tables appropriately in joins
· Utilize single pass reads – Read once and load multiple targets
· Minimize aggregation calls – Use sorted input wherever possible
· Reduce custom coding, stored procedures, scripting
· Add proper comments in the Mapping description
· Use environment variables for the services name in Unix scripts
For e.g-> pmcmd startworkflow –sv $INT_SVC –dn $DOMAIN_NAME ….
|
2
|
Lookup Transformation
|
· Reduce the dataset being cached - Override if required
· Remove un-used ports from the Lookup Transformation
· Place all the conditions with equality operator first
· If the number of calls can be reduced consider replacing with unconnected uncached lookup
· Consider using persistent cache for frequently accessed lookups
· When handling multiple matches, use “Return any matching value” setting whenever possible – Improves lookup performance
|
3
|
Aggregator Transformation
|
· Group by simple columns
· Use sorted Input
· Filter data before aggregation
· Limit port connection
· For high volume, increase the index and data cache size – avoid paging to disk
|
4
|
Operations and Transformation optimization
|
· Numeric operations are faster than string operations
· Operators are faster than functions (e.g. || vs CONCAT)
· Use Router transformation instead of multiple filter transformation
· Use Sorter transformation before an Aggregator transformation
· Use a Normalizer to pivot rows rather than multiple instance of the target
· While using Joiner transformation ensure that smallest source is set as the master
|
Wednesday, September 18, 2013
Informatica Power Center: Transformations and Optimization Best Practices
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment