Balancing of jobs
Posted: Wed Oct 22, 2008 9:11 pm
I'd like to know or get some advice from any shops or projects that anyone has worked on where there was a requirement to ensure that all records were accounted for during processing, i..e records input - rejects/dropped = records extracted. How was this done, i.e. data stage job, sql , scripts etc. If it has not been done, what things do you have in place to ensure processing integrity.
One of my customers requires it, but due to the complexity and number of different jobs there are alot of rejects (for business reasons, i.e. not necessarily data issues), remove duplicates, aggregations, creation of records etc. which make trying to create a generic formula for all jobs impractical. The main goal is to prove to auditors and downstream consumers of the data that there is processing integrity on the system as well as provide a report if needed.
Thanks,
One of my customers requires it, but due to the complexity and number of different jobs there are alot of rejects (for business reasons, i.e. not necessarily data issues), remove duplicates, aggregations, creation of records etc. which make trying to create a generic formula for all jobs impractical. The main goal is to prove to auditors and downstream consumers of the data that there is processing integrity on the system as well as provide a report if needed.
Thanks,