OpenStack DBaaS Trove: MySQL Replication Overview

May 10, 2024

Featuring Denis Makogon, Mirantis

Are you an OpenStack Cloud Database service user? Provisioning and use of database single instances not satisfying your application requirements due to data high load?

Trove is only able to deliver single instances of SQL/NoSQL databases for customers, but a database in single instance (SI) mode has a set of limitations dictated by the environment where it’s installed, primarily related to read/write operations. By using replication, we can deal with those limitations for read/write operations, so replication improves database performance for certain tasks.

To solve this problem, we can use replication at the MySQL database service level. This replication mechanism improves data availability and scalability, provides a fail-safe backup, and provides load spreading. You can use it to enhance the availability of data in different locations or split workload to different servers by distributing data by providing a mirror, merge data from different locations, provide a mirror for disaster recovery, or audit or analyze database usage.

This session will cover:
Trove replication design.
Mysql native replication schemes
Available replication topologies presets
Replication management API
Missing features
Once you’ve attended this talk, you will be familiar with Trove replication architecture, common use cases for given replication topology presets, and replications management in Trove. You will also be able to work with Trove as a database replication set provisioning tool for any back-end required applications.

source

by OpenInfra Foundation

openstack