1. When you join unsorted data using Joiner transformation, You can optimize performance by designating the source with fewer rows as master source.
2. When you join Sorted Data using a joiner transformation, you can optimize performance by designating the source with few duplicate key values as the master source.
3. You can edit a port in a mapping and then propagate changes to the other transformations,expressions, and conditions in the mapping.
4. Create an entry in ODBC.ini file in server to connect to a sybase IQ, SQL Server using ODBC.
5. If a lookup transformation has multiple conditions, you can improve performance by placing
conditions that use the equality operator first on the condition tab.
6. You can Optimize session performance by specifying the DB partitioning partition type for multi-node IBM DB2 targets.
7. You can improve the lookup performance by building the cache files concurrently.
8. you can send text message to all users connected to the repository by
selecting the repository in the administration console and selecting actions>Notify Users.
2. When you join Sorted Data using a joiner transformation, you can optimize performance by designating the source with few duplicate key values as the master source.
3. You can edit a port in a mapping and then propagate changes to the other transformations,expressions, and conditions in the mapping.
4. Create an entry in ODBC.ini file in server to connect to a sybase IQ, SQL Server using ODBC.
5. If a lookup transformation has multiple conditions, you can improve performance by placing
conditions that use the equality operator first on the condition tab.
6. You can Optimize session performance by specifying the DB partitioning partition type for multi-node IBM DB2 targets.
7. You can improve the lookup performance by building the cache files concurrently.
8. you can send text message to all users connected to the repository by
selecting the repository in the administration console and selecting actions>Notify Users.
The concept of Additional Concurrent pipelines and perform prebuilt lookup cache are different
Additional Concurrent pipelines:
Ans: In General a mapping starts build lookups when its start reading source data and at certain level the record
propagation happens and at the first record touches the lookup then the lookup building happens the lookup
building happens in sequence one after the other
When the Concurrent pipelines is selected in the session level (Default is Auto), then all the lookup builds start
parallel this will reduce the time load time
Note: This is not preferred as the space consuming will be more and changes of session failure is more
Prebuilt lookup caches:
Ans: By selecting this the session looks for the same lookups and for that I will only one cache and will be shared by the
other same lookups in the same mapping i.e if a mapping have test_table as lookup for 10 times with same condition then
mapping will be build cache for one lookup only
Note: Good practice to use this
No comments:
Post a Comment