Book Image

Learning PostgreSQL 10 - Second Edition

Book Image

Learning PostgreSQL 10 - Second Edition

Overview of this book

PostgreSQL is one of the most popular open source databases in the world, supporting the most advanced features included in SQL standards. This book will familiarize you with the latest features released in PostgreSQL 10. We’ll start with a thorough introduction to PostgreSQL and the new features introduced in PostgreSQL 10. We’ll cover the Data Definition Language (DDL) with an emphasis on PostgreSQL, and the common DDL commands supported by ANSI SQL. You’ll learn to create tables, define integrity constraints, build indexes, and set up views and other schema objects. Moving on, we’ll cover the concepts of Data Manipulation Language (DML) and PostgreSQL server-side programming capabilities using PL/pgSQL. We’ll also explore the NoSQL capabilities of PostgreSQL and connect to your PostgreSQL database to manipulate data objects. By the end of this book, you’ll have a thorough understanding of the basics of PostgreSQL 10 and will have the necessary skills to build efficient database solutions.
Table of Contents (23 chapters)
Title Page
Credits
About the Authors
About the Reviewers
www.PacktPub.com
Customer Feedback
Preface

Tuning performance for write


High write loads can have different patterns. For example, this can be a result of writing events to PostgreSQL or it can be a result of a bulk load from a database dump or an ETL job. One can tune PostgreSQL for high write loads by doing the following: 

  • Hardware configuration:
    • One should use RAID 1+0 instead of RAID 5 or 6. RAID 10 has much better performance for heavy writes. Also it is better to store transaction logs (pg_xlog) on a separate hard disk.
    • One can use SSD hard disks with Write-back cache (WBC), which significantly increases write performance. Also make sure your SSDs can persist cached data on power failure.
  • PostgreSQL server setting:
    • fsync: By default, fsync is on. This parameter makes sure that the database can be recovered in the event of a hardware crash. fsync makes sure that the data is actually written on the hard disk. You can disable this option if you can trust your hardware. Hardware failure might lead to corrupt data if this option is...