The "view data" functionality should not be used for viewing NLS data as it seldom does that correctly. I am fairly certain that your textfile data is not UTF-8; once you tell DataStage what the correct text file NLS mapping is, it will correctly convert to UTF-8.
If the source stage is set to UTF-8 then DataStage is reading the data correctly (despite the view-data output). What are your enviroment settings for DB2 and which DB2 stage are you using to write?
You cannot use the "view data" in DataStage to correctly view multibyte data, unfortunately.
Is your DB2 instance and/or table also set up to use UTF-8?
I am able to load the data into DB2 table using ODBC Enterprise stage. and able to view the data in DB2 table as it looks in source. Can any body let me know why its so. Why cannot i see the correct data when i use DB2 Stage