Book Image

PostgreSQL High Performance Cookbook

By : Chitij Chauhan, Dinesh Kumar
Book Image

PostgreSQL High Performance Cookbook

By: Chitij Chauhan, Dinesh Kumar

Overview of this book

PostgreSQL is one of the most powerful and easy to use database management systems. It has strong support from the community and is being actively developed with a new release every year. PostgreSQL supports the most advanced features included in SQL standards. It also provides NoSQL capabilities and very rich data types and extensions. All of this makes PostgreSQL a very attractive solution in software systems. If you run a database, you want it to perform well and you want to be able to secure it. As the world’s most advanced open source database, PostgreSQL has unique built-in ways to achieve these goals. This book will show you a multitude of ways to enhance your database’s performance and give you insights into measuring and optimizing a PostgreSQL database to achieve better performance. This book is your one-stop guide to elevate your PostgreSQL knowledge to the next level. First, you’ll get familiarized with essential developer/administrator concepts such as load balancing, connection pooling, and distributing connections to multiple nodes. Next, you will explore memory optimization techniques before exploring the security controls offered by PostgreSQL. Then, you will move on to the essential database/server monitoring and replication strategies with PostgreSQL. Finally, you will learn about query processing algorithms.
Table of Contents (19 chapters)
PostgreSQL High Performance Cookbook
Credits
About the Authors
About the Reviewers
www.PacktPub.com
Customer Feedback
Preface

Grouping


In this recipe, we will be discussing the optimizer node type, which will be chosen during the group by operation.

Getting ready

As we discussed group or aggregate operations in the previous recipe, grouping operations will have performed based on the group key list. The PostgreSQL optimizer chooses hash aggregate, when it finds enough memory and if not, group aggregate will be the option. Unlike hash aggregate, the group aggregate operation needs data to be sorted. If the group columns have a sorted index already, then group aggregate will choose over the hash aggregate as to reduce the memory usage.

How to do it…

  1. To demonstrate group aggregate, let's run the query in the benchmarksql database to get the count of customers, grouped by their city:

     benchmarksql=# EXPLAIN SELECT COUNT(*), c_city FROM
               bmsql_customer GROUP BY c_city;
                                                QUERY PLAN                                             
    ------------------------------------...