I have a partitioned table that contains approx. 16 billion rows. The table has a clustered columnstore index and one partition aligned non-clustered index. The partitions are all on the same filegroup. The table is partitioned by day. I plan to run a weekly process that truncates partitions containing data older than 90 days. The process also removes the partition (using a partition function merge range stmt). During a given weekly run, there may be approx. 500 million rows truncated across multiple partitions. Should I be rebuilding the indexes or updating stats after truncating that much data?
On SQL Server 2017