Continuing on the theme from part 1, we’ll have already differentiated between caching and statistics problems and found out why updating statistics often gives a false positive! But what if the real problem *is* statistics-related? In this session, I’ll show you some specific scenarios around which SQL Server statistics are incorrect! We’ll look at when sampling is an issue, inaccuracies in the histogram, and uneven distribution. Better yet, I’ll show you different techniques for making the problems go away!

Presented by Kimberly L. Tripp at SQLBits XX