sql server - Performace and Sizes of Non-Clustered Indexes drops as size of clustering key increases? -

Excerpt from

The width of the clustering is not required, however, only the affected package index Clustering keys, being aware of rows, are located in each non-clustered index. Therefore, a detailed clustering key increases the size of all non-clustered indexes, and also reduces their efficiency.

Why should the databases increase in the non-clustered index size of the pointer if sizeof ( Int *) is the same as the size (four *) or is there any other addressing mechanism I do not know about?

Pointer does not point to data Page: This indicates the index index. Micro, but there is no clustering index in which each NC index indicates Row ID (RID) in entry data pages.

So if you have char (10) as your key, every NC indicator entry has 10 byte pointer if you have an integer, then this is 4 bytes.

For four vs. int, there are other ideas: at least matching (which determines the sensitivity of case, pronunciation, kana and width).

And you apply the string of 4 bytes only to the photo char (4) . nchar (4) 8 bytes, varchar (4) 2 to 6 bytes (2 byte length), nerve (4) 2 to 10 Bytes