DSXchange

The job has 2 odbc stages and in one of the odbc stage I used a query which has just 2 conditions using where.
The issue is the job takes 70 minutes to run. Is there any way to increase the performance.I see in the monitor of that job, seems the odbc with the query takes long time.

Some additional information such as how many records the table has, the many records the output SQL produces, how many columns you are trying to extract, if the columns you use in the where conditions are indexed or not will help us answer you better.

Hi,
it's really hard to give you an advice without knowing any details

e.g.
- the job requirements (number of records to be processed, selectivity of the where clause, ...)
- the design (downstream stages, job logic, ...)
- configuration of your system (degree of parallelism, environment, setup, used dbms...)

according to your statement that you are using a where clause in one of your odbc stages my first guess is that there is an index missing and your dbms has to perform expensive table-scans.


BR

Werner

sorry for the incomplete information

The ODBC stage is used and the DB is informix

the records are around 700000

A query is used to get data from 2 tables and with where condition

for two conditions.

like the query is

select a.name,a.phone,a.fax,b.state from person_info a,address b where a.pid=b.pid and b.status="C";

Are there indexes on person_info.pid, address.pid and address.status?