Answer / sreedhar lokaray

Many people are not sure how exactly to tune the
performance of a mapping. There is no thumb rule for
performance tuning. A tuning done in one mapping cannot be
applied into another mapping. But still below given are
some of the performance tuning tips.

Joiner instead of Lookup: If the lookup table has large no.
of records then the performance will affect because it has
to build the cache. Instead use a joiner transformation and
check it as detail rows. This will avoid building cache.

Sorted Input: Use sorter transformation before joiner or
aggregator. This will improve the performance.

Source Qualifier: Try to filter out unwanted records in the
source qualifier before moving to the next transformation.
Also delete unconnected/unused ports.

Indexes: Try to have proper indexes on the required fields
of the tables in the database which will increase the
performance. At the same time having indexes on the target
will kill the performance when you try to insert huge
volumes of data. To avoid this, disable the indexes on the
target (using sql loader direct load process) and then load
the data. After loading enable the index back.

Keep in mind that tuning performance of one transformation
can kill the performance of other transformation. We have
to maintain the equilibrium.

You have defined the following: - Commit Type = 'Target' - Commit Interval = 10000 - writer buffer block can hold multiple 7,500 rows - you are loading 40,000 records into the target After how many records will the Informatica Server issue commit commands? a)7500, 15000, 22500, 30000, 40000 b)15000, 22500, 30000, 37500, 40000 c)15000, 22500, 30000, 40000 d)15000, 30000, 40000

7 Answers   Deloitte,

---

Write the program through which the records can be updated?

0 Answers  

---

Scenario:-  Below is the requirement. Source:-  NAME          ID                    Requirement  RAVI            1                      (no need to repeat as it ID is 1) KUMAR        3                     (repeat 3 times as it ID is 3) John             4                     (repeat 4 times as it ID is 4) Required Out Put:- Name                   ID RAVI                     1                    KUMAR                3      KUMAR                3 KUMAR                3  John                     4 John                     4 John                     4 John                     4 Scenario 2:- Source Data  ID              NAME 1,2            NETEZZA,ORACLE 3,4,5         SQL Server, DB2, Teradata Required Output:-  ID                   NAME 1                  NETEZZA 1                  ORACLE 3                  SQL Server 3                  DB2 3                  Teradata

1 Answers   Cognizant,

---

what is the use of Java Transformation in Informatica 8.x?

2 Answers   TCS,

---

What is transformation?

0 Answers  

---

What will be the approach?

0 Answers  

---

Difference Between ETL & ELT?

1 Answers   TCS,

---

What is a design?

1 Answers  

---

difference between source based commit? and target based commit? which is better with respect to performance?

6 Answers   IBM,

---

Can anyone tell me, when a flat file contains different kinds of delimeters available for each line ( For Ex: First line contains [,*&#$] and second line contains [%@$#%^]. Like that all records having different kinds of delimeters. How can we implement this kind of flat file in informatica?

2 Answers  

---

There are 3 groups in router transformation in which there are no conditions mentioned, so what happens if I connect these groups to the 3 targets? How many records gets into each target.(suppose if we are getting 10 records from the source)

4 Answers  

---

Hey I am net to informatica? Can any one explain me step by step How scd will work ?

1 Answers   Infosys,

---