May 9, 2012 at 5:12 pm
Hi All,
Performance wise is there any difference using (by non-clustered index) ROW ID(for heap) vs Clustered Key id?
Thank You
May 9, 2012 at 5:34 pm
I believe the answer is Yes there is a performance difference.
The non-clustered index on a HEAP must find the page using the ROWID then find the row data on the page. Clustered table the non-clustered index points to the exact row in the clustered index by the clustering key. The linked-list nature of the B-tree also increases the probability that related rows will be in memory.
May 9, 2012 at 5:47 pm
Correct me if i am wrong:
Using Clustering key sql server retrives data from memory vs
using Rowid from disk
May 9, 2012 at 9:28 pm
Depends if the data page needed for the query is in memory (cached) or not.
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply