When to use which stage in Datastage?

Copy STAGE-To drop a Particular column

Sort STAGE-sorting,generating Key change and similar to order by clause in oracle

Filter STAGE-Similar to where clause in oracle but we can not perform Join operation

Lookup,Join,Merge-To perform Join operation

Pivot Enterprise STAGE-Rows to columns and columns to Rows

External Filter STAGE-Filter the records by using Unix filter commands like Grep etc

MODIFY STAGE-Metadata conversion,Null Handling and similar to conversion functions in oracle

FUNNEL STAGE -Combining the multiple input data into a single output.Metadata should be same for all the inputs

REMOVE DUPLICATES STAGE-To  remove  duplicate  values  from  a single sorted  input.

ENCODE / DECODE STAGES:To encode/compress a data set using UNIX encoding  commands like gzip etc

TRANSFORMER STAGE:
a)Filtering the Data(constraints)
b)Metadata conversion(Using Functions)
c)Rows to columns and columns to Rows(Using Stage variables)
d)Looping
e)Creating a counter(Using macros)-Counter using Transformer

SURROGATE KEY GENERATOR STAGE-To generate SURROGATE KEYs similar to oracle Database sequence

Aggregator Stage:To perform Group by Operations like max,min etc similar to Group by clause in oracle

ROW GENERATOR STAGE:To generate a set of mock data fitting the specified metadata when no real data is available

XML OUTPUT STAGE -To convert tabular data such as tables and sequential files to XML hierarchical structures.

SWITCH STAGE-  It performs an operation similar to  the  switch  statement  in  C and to filter the data

CHANGE CAPTURE STAGE-To identify Delta changes(inserts,updates,deletes etc) between two sources

oracle connector-To connect to the oracle Database.