Two longitudinal table a summary (200 w), a subsidiary (7000 w), they need to link into a horizontal table, horizontal table column is about more than 150 columns.
So I write a lot of sum (decode), performance has a problem, now for optimization solution

CodePudding user response:

Demand is not clear,
Is a one-time conversion is complete, or timing refresh
Would you please consider using materialized views

CodePudding user response:

We this two tables for incremental data every month (summary: hundreds of thousands of, detail: tens of millions). The logic is implemented in the PKG. Then use etl scheduling. A month. But now need a full amount of script for the first time. So it will make a great amount of data

CodePudding user response:

11 g? There is a corresponding implementation pivot, unpivot

CodePudding user response:

Even if the full amount for the first time, you can also use the job in batches of a career change,