Row Goals Gone Rogue

  This post discusses “row goals“, but with a twist. The point is to illustrate how row goals can cause unnecessarily slow queries. First, run this script: USE tempdb GO IF OBJECT_ID (‘even’) IS NOT NULL DROP TABLE even; IF OBJECT_ID (‘odd’) IS NOT NULL DROP TABLE odd; GO CREATE TABLE even (c1 int, c2…

1

Query Tuning Fundamentals: Density, Predicates, Selectivity, and Cardinality

A few days ago I was in the middle of writing up a quick post about a query performance problem I recently ran into. The writeup referenced predicate selectivity, and I found myself wondering whether everyone who came across the post would have a good intuitive understanding of what that referred to. Just in case,…

7

Living with SQL’s 900 Byte Index Key Length Limit

We recently had a situation where we needed to interface with an external non-relational data source that happened to use really long strings to identify entity instances. These identity strings had theoretically unbounded length, but in practice they were never more than 1000-2000 characters. The database layer needed to accept an identifier of this sort…

5

Query Hash Statistics, a query cost analysis tool, now available for download

Some time ago I described the query fingerprint and query plan fingerprint (a.k.a. query hash / query plan hash) features that were added in SQL Server 2008.  A fingerprint identifies all queries or all plans that have the same “shape”.  With query fingerprints you can get the cumulative cost of all executions of a query…

1

Purging Data

This post is about deleting a lot of rows, as you might do in a data archiving or purging task.  By “a lot of rows”, I mean anything from a few thousand rows up to billions.  This may seem elementary, but there are some surprising problems that can appear if you use a simple DELETE…

4

Finding procedure cache bloat

Explicitly parameterizing queries is a well-known best-practice for database app developers.  There are cases where it can make sense to execute a query unparameterized, but, in general, you should default to explicit parameterization.  It can make your queries faster by avoiding unnecessary compilation when the “same” query is run repeatedly with different parameters, and it can reduce…

1

Sometimes the Simplest Solution Isn’t the Best Solution (The Optional Parameter Problem)

Programmers should naturally gravitate toward the simplest, most elegant solution.  This is because the simplest coding solution is so often the best solution: simple solutions are cheaper to implement; easier for others to understand, maintain, and extend; and less prone to bugs.  Simplicity can also be associated with efficient and consistent execution, but performance is…

11

Defining Complex Server "Health" Policies in SQL 2008

Policy Based Management (PBM) is a new feature in SQL Server 2008 that allows you to define a set of policies that capture the intended state for a group of servers.  For example, you could define a policy that says that your user databases should all have the auto update statistics database option enabled.  (If…

7

Query Fingerprints and Plan Fingerprints (The Best SQL 2008 Feature That You’ve Never Heard Of)

  In versions of SQL Server before SQL Server 2008, it can be difficult to determine the cumulative cost of the queries running on a server if the workload includes unparameterized queries.  The only truly reliable method is to capture a Profiler trace during a representative time period, then post-process the trace with a utility…

4

Are you using SQL’s Missing Index DMVs?

Did you know that your SQL Server is keeping track of the indexes that it thinks you should create?  The “missing index” DMVs in SQL are a really great new feature in SQL Server 2005 that (in my opinion) seem to have been underutilized so far.  If you want to see if this feature can spare you…

29