This is the first part of a series on quorum queues, our new replicated queue type. We’ll be covering everything from what quorum queues are, to hardware requirements, migration from mirrored queues and best practices.
Mirrored queues, also known as HA queues have been the de facto option for years when requiring extra data safety guarantees for your messages. Quorum queues are the next generation of replicated queue that aim to replace most use cases for mirrored queues and are available from the 3.8 release and onward.
In this blog series we’re going to cover the following:
- Part 1 - Understand the need for a new replicated queue type and how it works (this post)
- Part 2 - Why your choice of storage drive matters with quorum queues
- Part 3 - Quorum queues and flow control. Concepts.
- Part 4 - Quorum queues and flow control. Single Queue Benchmarks.
- Part 5 - Quorum queues and flow control. Stress Tests.
- Part 6 - Quorum queues and cluster/VM Sizing (coming soon)
- Part 7 - When and how to migrate from mirrored to quorum queues (coming soon)
- Part 8 - Best practices and gotchas (coming soon)
- Part 9 - Monitoring and admin operations (coming soon)