November 11, 2015 at 2:45 am
Hi,
I am having trouble finding information on how I should handle statistics on clustered columnstore indexes. I am currently using a partitioned clustered columnstore index, and I am planning to rebuild the partitions that have had data loaded into them. If I were to rebuild indexes within a rowstore table, I would potentially want to manually rebuild statistics for the column statistics (the index statistics would already be rebuilt). Is there anything equivalent I need to factor in, for a columnstore index? Or would a rebuild of the partition be sufficient?
November 11, 2015 at 6:01 am
WWOHD.
What would Ola Hallengren do?
I think it would be a good idea to get the index optimise script from Ola.
Using that, you can run the index optimise script with logging only and no execute to see what Ola's script would do, as it will log all the commands to table.
I am guessing it would automatically try and update statistics on table on partition, for those partitions where there was a change to the data.
Just take note that since it is not executing, you may see it try and rebuild an index and update the stats for that index.
In real life, it will check the stats after the rebuild.
So it will not rebuild the stats if the stats is a child of the index which has been rebuilt and hence new.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply