Answer / mike

The goal of performance tuning is optimizing the session
performance. so session runs during available load window
for the informatica server. To increase the session
performance we need to follow the below steps
1)The performance of the informatica server is related to
the network connections. Generally data moves across the
network at less than 1 mb per second, where as local disk
moves five to twenty times faster. so network connections
aften affect the session performance.Hence avoiding the
network connections session performance gets increased.
2)If your flat files are stored in a machine other than
informatica server, need to move the files where
informatica server exists.
3)By minimizing connection to sources, targets and
informatica server session performance gets increased.
4)By running multiple informatica servers against same
repository, thus distributing the session load on multiple
servers may increase session performance.
5)By running the informatica server in ASCII data movement
mode session performance gets improves this is because,
ASCII mode stores a character value in one byte where as
UNICODE mode stores 2 bytes per character.
6)By avoiding transformation errors session performance
gets increased.
7)If a session having aggregator t/r, we can use
incremental aggregation to improve session performance.
8)If a session having lookup t/r , by enabling lookup
chache session performance gets increased.
9)aggregator, rank, joiner t/r's will always slows down the
data load this is because they must group the data before
processing it. In such case by using sorted input option we
can improve session performance.
10)If your target contains key constraints and indexes, it
may slow down the data load. In such case need to drop the
constraints and indexes and rebild them after completing
the session....

If something was taking much time need to check the
bottlenecks starting from
source,target,transformation,mapping,session,workflow

can we generate reports in informatica ? How?

3 Answers  

---

How can i send first half of the records to one target and Remaining to other target?

10 Answers   TCS,

---

can anyone explain me about retail domain project in informatica?

0 Answers   Zensar,

---

How to find from a source which has 10,000 records, find the average between 500th to 600th record?

3 Answers  

---

my source has 2 columns. 1st column data: ABC ABC ABC XYZ and 2nd column data: 1 2 3 4. And my target should be ABC 1 2 3, XYZ 4

2 Answers   Zensar,

---

Differentiate between mapping parameter and mapping variable?

1 Answers  

---

what is shared Cache. when we will use shared Cache?

4 Answers  

---

how do u get the first record from 50,000 records ?

10 Answers   TCS, UBS,

---

in u flatfile some of fact records are missed then u load the diminision records are not if u load diminsion records what about fact table records

0 Answers  

---

Hi experts, > I 'm having data like this and it's in flat file. > sno,name,marks,result > 1, ak, 80,p > 2, , ,20,f > 3,jack,55,p > > now in the second row a comma(,) is placed instead of name(char). > > when importing into SQ , in flat file wizard the num of columns increased to 5 instead of 4. > > could you tell me how to over come this issue.........

1 Answers   TCS,

---

How to display last 5 records in a table ? With out Top key word and doing order by desc Advance thanks

2 Answers   Puma,

---

What is router transformation

4 Answers   CSC,

---