We know some of you are eager to learn what’s happening on the backend. Today, we’d like to shed some light on the recent changes we made to our database, why we made the upgrade and the consequences for Pax Dei.
Overview
To improve the reliability and scalability of our system, we recently migrated from our legacy Amazon RDS Postgres system to Amazon Aurora (Postgres compatible). This migration required us to schedule two maintenance breaks to make necessary database changes with a double goal.
- Enhanced Reliability:[list]
- Our previous setup with Amazon RDS Postgres, which used a primary and a secondary node, offered a reliable setup where the secondary node can take over and be promoted to a master node in the case of a failure. This failover worked well but could result in several minutes of downtime during the switch.
- Aurora databases have a shared, reliable storage layer between all nodes, which significantly reduces recovery time from minutes to seconds by offering rapid failover from a primary to a secondary node.
- Aurora also allows us to dynamically add and remove additional nodes from the database cluster.
- The old system required significant downtime to implement changes - as any major changes require us to use the failover mechanisms between nodes while performing upgrades or changes to each.
- Aurora provides the flexibility to scale resources up or down and adjust database node sizes with minimal disruption.
[/list]
With the migration to Amazon Aurora now complete, we now have a better uptime and enhanced reliability with faster failover, as well as seamless scalability to support future growth.
This migration positions us to better serve you, our community, by providing a more robust and flexible database infrastructure.
Stay tuned for more insights on Pax Dei development!
Pax vobiscum,
The Mainframe Team
