DSXchange

I am analyzing the table daily before UPSERT and after UPSERT so STATS will be updated daily. And my job design is OracleEnterpriseStage--------->Transformer----------->OracleEnterpriseStage There are no transformations in INSERT and UPDATE, they are straight SQL queries. My Initial load will be 70...

Hello dav_mcnair,

I am looking for a similar solution - could you share your solution if you have come up with something

thanks
tsanan

The job works fine and I get the right results when I run all the stages in sequential mode. But when I partition the transformer in Hash partition and sort on the acctid and calid (I also tried sorting on acctid, calid in the SQL statement and Hash partitioning just on acctid as well) I do not get...

How are the data partitioned on both inputs to the Lookup stage? I had been trying different ways: 1. Used Order by actid,calid clause in the enterprise stage which is a stream input to the lookup stage. The reference lookup being Parallel execution mode and Preserve partition set to Propogate. The...

Thanks for your response Ray. What's the key of the reference table for the lookup? What partitioning algorithm are you using for the reference input link. The key to the lookup is calendarid which is a surrogate id and am looking up for the calendar year and month. I suspect you may be missing some...

Hello All I am having a problem calculating values based on previous row values while Hash partitioning. I would appreciate any advise. I have account-wise daily data and am trying to create monthly aggregates: acctid, calid, val1, val2,val3,closing - this is the layout of my data there are roughly ...

hi,
i am very new to custom build stages, so where can i get this guide