Whitepaper: Highly Available Postgres Clusters
This whitepaper discusses the scenarios when designing a high availability architecture for PostgreSQL, including failover and switchover. It outlines strategies for backup and disaster recovery and proposes an architecture that relies on best practices for high levels of reliability in production.
Topics covered in this whitepaper:
A solution that comes with the built-in ability to formulate a backup strategy and recover from disasters. This solution also provides the ability for Point-in-Time-Recovery.
A solution that comes with rolling upgrade capabilities, allowing you to upgrade individual nodes sequentially until the entire cluster is on the required version.
A solution that comes with the ability to handle planned switchovers or unplanned failures. It even supplies redundancy provisions in case of complete data center failures.