Background Several weeks ago, I ran some tests on the new implementation of Unicode Compression in SQL Server 2008 R2 - both space-related and performance-related.  Of course it was pointed out that my performance tests were somewhat flawed, because I was using data from a very wide and atypical variety of characters sets.  I also ...

When SQL Server 2008 R2 was first released, I wrote a quick blog post discussing the additional space savings gained by Unicode compression. In one of the follow-up comments, Adam Machanic was quick to suggest that compression can bring significant CPU overhead, and can as much as double the amount of time it takes to retrieve data.  He was ...