Rebuild old partitions for a new field

[BillColeman]

I have incremental refresh configured on a large fact table that has been in production for a while, but I'll need to add a new field to the table now. I can redefine the table to include the field using ALM toolkit, but the data only gets populated in the new field for the current partition. Is there a way to force a complete refresh/rebuild of older partitions so that the new field's data can be populated retrospectively?

The only think I can think of is to remove the partitioning altogether, do a full refresh and then replace the partitioning. I really don't like the idea of doing this on a production dataset. There has to be a cleaner solution to rebuild old partitions.

I did try via SSMS - connecting to the dataset as if it were an AAS cube and selecting & rebuilding partitions, but this does not work. Anyone else encountered this puzzle?

[ibarrau · Accepted solution]

Hi. I was thinking about the same as you. Deleting and refreshing all over again. However, guy in a cube has recently published an interesting appproach that might help.

https://www.youtube.com/watch?v=5AWt6ijJG94

Hope that helps,

[Victor_Yang]

HI,

Did you find a way to add new field to all partitions wthout deleting the partition?

Thanks

[BillColeman]

Ah man, I was so close! I just had partition refresh process a little wrong from SSMS. Thanks for the link to that video. My trick bag has been updated 🙂