Question

What is “better” generating Primary Keys on the database or generating them in application code, specifically when using GUID/UniqueIdentifier datatype for the keys.

I have read up on the difference between using Guid’s and int data types, and it sounds like Guids are feasable for so called “generating offline”.

E.g.

instead of having a NEWID() contstraint in the database, In one project (where we are using Entity Framework) we use in the application code Guid.NewGuid() to generate the PK when inserting data.

Is this a bad approach ?

My concerns are:

• Database indexes: Database performance because Id’s might not be sequential
• The one in 64 billion chance that the key is already used. (considering that the application will not be enormous but may need room to grow)
• Perhaps there are other disadvantages ?

Answer 1

well actually, GUID could be sequential from SQL Server 2005. There is a function in named NEWSEQUENTIALID() , link here

Creates a GUID that is greater than any GUID previously generated by
this function on a specified computer since Windows was started. After
restarting Windows, the GUID can start again from a lower range, but
is still globally unique. When a GUID column is used as a row
identifier, using NEWSEQUENTIALID can be faster than using the NEWID
function. This is because the NEWID function causes random activity
and uses fewer cached data pages. Using NEWSEQUENTIALID also helps to
completely fill the data and index pages.