Book Image

PostgreSQL Development Essentials

By : Baji Shaik
Book Image

PostgreSQL Development Essentials

By: Baji Shaik

Overview of this book

PostgreSQL is the most advanced open source database in the world. It is easy to install, configure, and maintain by following the documentation; however, it’s difficult to develop applications using programming languages and design databases accordingly. This book is what you need to get the most out of PostgreSQL You will begin with advanced SQL topics such as views, materialized views, and cursors, and learn about performing data type conversions. You will then perform trigger operations and use trigger functions in PostgreSQL. Next we walk through data modeling, normalization concepts, and the effect of transactions and locking on the database. The next half of the book covers the types of indexes, constrains, and the concepts of table partitioning, as well as the different mechanisms and approaches available to write efficient queries or code. Later, we explore PostgreSQL Extensions and Large Object Support in PostgreSQL. Finally, you will perform database operations in PostgreSQL using PHP and Java. By the end of this book, you will have mastered all the aspects of PostgreSQL development. You will be able to build efficient enterprise-grade applications with PostgreSQL by making use of these concepts
Table of Contents (17 chapters)
PostgreSQL Development Essentials
Credits
About the Authors
About the Reviewers
www.PacktPub.com
Preface

Query tuning


Before discussing query tuning, first we should know how to write a query, and then we can tune it later on. Queries can be written in multiple ways, as per the requirement; however, we have to prefer an optimal way to write queries.

For example, say that your requirement is to get the row count from a static table. You can achieve this through the following three methods:

Method 1:

postgres=# SELECT SUM(1) FROM customers;
sum
------
1000
(1 row)
Time: 4.308 ms

Method 2:

postgres=# SELECT COUNT(*) FROM customers;
count
-------
1000
(1 row)
Time: 3.128 ms

Method 3:

postgres=# SELECT reltuples FROM pg_class WHERE relname =  'customers';
reltuples 
-----------
1000
(1 row)
Time: 0.576 ms

From the preceding three possible methods, we see that Method 3 gives the best result, and Method 2 is optimal when compared to Method 1. Method 1 takes a bit more time, when compared with Method 2, since it needs to SUM all the records at once, whereas the COUNT aggregate follows a different...