We have a job with Oracle Enterprise stage as the target. The job has 1.3 million+ records coming from source. This job has been migrated from DatStage7.5.1 The design is as below:
Code: Select all
ODBC--->Transformer-->Remove Duplicates-->Oracle Enterprise
|
|
Sequential file
But in DataStage8.1, it takes 40 minutes to process 0.18 million records. The write method in the Oracle Enterprise stage is Upsert(User defined Update & Insert)
So can anyone help us understand why does the same job takes very long time to run and finally hangs in DataStage 8.1?