The ability to create and update statistics is a crucial liberty for the query optimizer. The optimizer needs to know how it is going to query underlying tables, and when the statistics do not match the data, it is very probable that it will choose a non-efficient method for querying. I’ve seen and heard a [...]
Archive of posts filed under the SQL Optimization category.
Execution Plans
The execution plans SQL Server creates and uses is a huge topic with many intricacies. Now I have never spoken to anyone on the SQL Server Development team, however I would imagine that there are some extremely sharp people in the query optimization team. In a nutshell, this is where the magic happens in SQL [...]
Understanding Batch Requests / sec
SQL Server’s Batch Requests represents the number of SQL Statements that are being executed per second. In my opinion this is the one metric that gives the best baseline on how your server is performing. It cannot be looked at by itself however. It is necessary to correlate with other metrics (especially CPU usage) in [...]
Understanding SQL Server Deadlocks
In order to fix deadlocks, it’s critical to understand why they occur. I’ve gotten some push back sometimes from readers when I write articles that do not give cookie cutter solutions, but the “teach a man to fish” adage couldn’t be more true for deadlocks. All of it revolves around locking, which is a central [...]
Comparing Query Performance
If you write TSQL then comparing the performance of two queries will probably be something you do on a daily basis. The difficult part of comparing queries is getting an accurate baseline. Why is this you ask? Because SQL Server has behind the scenes functionality that optimizes queries for multiple executions. This allows minimal resources [...]
SQL Server Join Algorithms
If you read execution plans enough, you’ve probably realized that when SQL Server joins tables together, it uses different internal algorithms. The three algorithms are: Loop Join Merge Join Hash Join These alogorithms that are used are based upon factors of the underlying data. Merge Join For the most part, this is the most efficient [...]
What is Parallelism?
Parallelism is a feature in SQL Server which allows expensive queries to utilize more threads in order to complete quicker. The query optimizer makes the determination of how expensive a query is based upon the Cost Threshold for Parallelism setting set at the SQL Server Instance level. This cost threshold for parallelism is an estimate [...]
Configuring Parallelism
Parallelism is a powerful feature in SQL Server designed to allow greater bandwidth for high impact queries. Most of the time however, I’ve seen too many queries qualifying for parallelism and essentially bringing the SQL Server box to it’s knees. To understand parallelism, we need to understand SPIDs (Server Process ID’s). These SPID’s act essentially [...]
Finding the Slowest Query using the Execution Plan
If you’ve stumbled upon this post, then you’re likely at a point where you’ve got a crap query that is slowing you down. This article tells you how to find that one slow query, within a procedure. You also probably know that the execution plan is what you need to use to find out how [...]
How to Optimize a Stored Procedure using the Execution Plan
Find the most costly statements Determine why the statement is costly Get an accurate baseline for the procedure Optimize Find the Most Costly Statments There are two main ways to determine costly queries. One is to execute the procedure including an Actual Execution Plan. The other is to get a time difference of a before [...]
