We are in process of developing a project which will process around 51 000 000 records daily (~around 500-600 inserts per second). We choose "Always-On" solution which will contain:

  1. One primary server
  2. Three replicas which has to work in synchronous mode

One of the main tables (45 columns, several datatypes) will be partitioned (on a smallint column).

My question is:

unique constraint and primary key on partitioned table has to also include partition column.

Which approach will be better:

  1. Create a clustered index on primary key (which includes partition column and ID column which is IDENTITY(1,1))
  2. Create a clustered index only on ID column (which is IDENTITY (1,1))

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.