Independent Flows within Same Parallel Job

jerome_rajan · Post by **jerome_rajan** » Mon May 21, 2012 10:04 pm

Was just wondering if it is possible to have independent flows within the same parallel job. Something like

SEQ --- TXN ----DS

DS --- CP ----ORA

Both these flows would be within the same job. I understand that this is possible but what would the order of execution be in such a case.

Also, I would like to know if any one has encountered a requirement where you had to employ such a design.

Thanks in advance

jwiles · Post by **jwiles** » Mon May 21, 2012 11:27 pm

Order of execution: The two streams will execute at the same time as they are in the same job.

Regards,

jerome_rajan · Post by **jerome_rajan** » Mon May 21, 2012 11:39 pm

I created a job that looks something like below

Code: Select all


                               SK1

        ROWGEN ----> SK2 -----> DS

SK1 creates the state file while SK2 generates the values. If like you said, both flows are executed at the same time, shouldn't the job have aborted due to the absence of the state file? How is it that everytime I run, the Sk1 creates the file first. Is it just a co-incidence?

ray.wurlod · Post by **ray.wurlod** » Tue May 22, 2012 3:49 am

SK1 and ROWGEN start at the same time. SK2 can't start till the first buffer of rows arrives from ROWGEN. Not a coincidence at all.

rameshrr3 · Post by **rameshrr3** » Tue May 22, 2012 8:42 am

I have used 2 independent flows in a job design , sometimes with a common lookup data source, I usually see that job monitoring in director or performance stats in designer is disabled in these cases . Is that a known issue / 'feature'