How can we help you today? How can we help you today?

Why recreate instead of rename?

Dear Support,

i have a question regarding the generated syncronizing script.

I have an example: 
left side: CREATE NONCLUSTERED INDEX [Name_Vorname] ON [dbo].[Inter] ([Name], [Vorname])

right side: CREATE NONCLUSTERED INDEX [Name] ON [dbo].[Inter] ([Name], [Vorname])

The resulting synchronisation script does the following: 
DROP INDEX [Name] ON [dbo].[Inter]
CREATE NONCLUSTERED INDEX [Name_Vorname] ON [dbo].[Inter] ([Name], [Vorname])

My question is: Wouldn't it be wiser (and a lot faster) if the alteration script only renames the index as all other parameters are equal?

It's kind of a showstopper to us as we need to synchronise very large tables and recreating doesn't qualify for the perfect solution.
Fenske
0

Comments

2 comments

Sort by
  • James B
    Thanks for your post. This same question was actually asked back in 2005 (ref. SC-1308) and the request was closed, with a comment of:
    No plans to do this

    While indeed it would be more efficient from SQL Server's point of view, because of sp_rename it would confuse system tables, and it would NOT be more efficient in SQL Compare (indeed it would be much more inefficient)

    I'm not sure of any more detailed reason as to why it would be more inefficient in SQL Compare unfortunately as that's all that was written on the request, but it sounds like performing the rename could actually make the problem worse and we have no plans to fix it.
    James B
    0
  • Fenske
    Thank you for your reply.

    With kind regards,
    Fenske
    Fenske
    0

Add comment

Please sign in to leave a comment.