Home/ Questions/Q 142523
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-05-11T07:57:01+00:00

I’m working on a table design that could involve many NULL values in about

  • 0

I’m working on a table design that could involve many NULL values in about 10 fields maybe 75% of the time the fields would be unused.

I just generated some fake data (a million records) and could not sense any impact on SQL Server 2005. Size difference was in the KB. Performance – no measurable difference after adding an index to the 3 non-nullable columns.

I know SQL Server 2008 has the sparse columns feature (which I assume is going to be used on the next SharePoint’s UserData table). I want my code to work on 2005 though. But lots of NULL values exist in the design of the current SharePoint UserData table. So if its good enough for Microsoft…

Any good articles, links, white papers on the drawbacks or pain points around many NULL values in SQL Server table? Anyone have any experience on what happens as you scale to 10 mil or 100 mil records?

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. I have never had a problem with the performance on multiple null columns, even on databases in the 100s of gigs size. I imagine you can end up with issues if you are running indexes on these fields and then using null in the query, but I have not seen this as a problem personally. Then again, I have not created database tables where every field except 3 was nullable.

    On the other hand, I see an architecture problem when most of the data is null. the general reason is either a) an improperly normalized database or b) an attempt to allow users to stage data in the end table rather than creating separate tables to ‘build’ data prior to committing to the database.

    It is up to you to determine the best architecture of your database.

    • 0