To get familiar with AWS Aurora, take a look at the recent AWS April 2016 Webinar Series – Migrating your Databases to Aurora and the June 2016 Webinar Series – Amazon Aurora Deep Dive – Optimizing Database Performance session lead by Puneet Agarwal.
- Faster recovery from instance failure (5x times or more vs. MySQL)
- Consistent lower impact on the Primary replica
- Needs additional throughput (theoretically 5x times for the same resources vs. MySQL). This was achieved by decoupling the cache and storage sub-systems and spreading them across many nodes as well as performing log commit first while DB manipulation is done asynchronously.
- Uses or can migrate to MySQL 5.6
- Comfortable with the Aurora I/O mechanism (16K for read, 4K for write, all can be batched if smaller)
- Gets more replicas (maximum of 15 vs. 5 in MySQL)
- Prioritizes recovery replica targets and sets replicas with different size than the master
- Needs virtually no replication lag – since replica nodes share the same storage as the master uses
- Able to decide about encryption at rest, at DB creation
- Accepts working with the InnoDB engine alone
- Eliminates the need for cache warming
- Only an additional 20% pricing increase over MySQL to gain all the benefits above
Originally posted by Jack Bezalel on his blog here.