When you are first developing a SQL Server system one of the last things that most people think about is performance. That unfortunately becomes one of the biggest problems that the developer has when the system goes into production. You end up finding out when someone calls and says the system seems to be running slow. It is these kinds of problems that are hard to find!
That is, unless you know how to diagnose the problems systematically. This is the goal of SQL Server 2008 Query Performance Tuning Distilled. This book provides the tools that you will need to find the problems. It does this by showing you how to set up mechanisms for collecting performance data so you can get to the heart of the problem. The book is 600 pages divided into 16 chapters.
Chapter One, "SQL Query Performance Tuning," begins by looking at what performance tuning is. You will look at how to establish a baseline, find bottlenecks, resolving problems, and then how to measure the improvements. Chapter Two, "System Performance Analysis," starts off by monitoring the system upon which SQL Server runs – namely Windows. With Performance Monitor you will learn how to use the different performance counters that are required to create a baseline.
Chapter Three, "SQL Query Performance Analysis," now examines the best methods to look under the hood of SQL Server and see what queries are being run on the system. This chapter provides a detailed look at SQL Profiler and Management Studio tools. Chapter Four, "Index Analysis," next looks at indexes and how they are built. You will see the difference between clustered and non-clustered and which works best with each type of query. The right index on the right column may mean everything to proper performance.
Chapter Five, "Database Engine Tuning Advisor," is a tool that helps identify an optimal set of indexes and statistics for a given workload without requiring an expert understanding of the database schema. Here you will learn how the tool works, how best to use it, and what its limitations are. Chapter Six, "Bookmark Lookup Analysis," is all about key lookup and the major overhead that is associated with non-clustered indexes. It is here that you will learn how to avoid these costs.
Article comments