I am working on a Data warehouse. I have tables with up to 200M records. Some of these tables have around 20+ indexes (I can't provide a reason why they have been created in the first place). This is making the job of maintaining these indexes too painful and has a direct impact on the DWH import job in both performance and run time.
How can I find the least used indexes on each table? (in order to get rid of them)