Member-only story
1 Fine Detail That Makes Amazon Aurora a Great SQL Solution
2 separated layers make a big difference
Layers on a cake are nice, right?
You could say the same about your database solution. Generally with a SQL database, you will have a layer for compute and a layer for storage. However, they will likely be physically dependent on each other since they would be on the same server
Because of this, you become more limited in regards to your options to configure, manage, backup, and scale your database server.
This is where Amazon Aurora comes into play, as the fine detail I alluded earlier is that Aurora is a SQL based solution that separates the compute and storage layers. It currently has full support for PostgreSQL and MySQL engines.
As a high level overview, the idea is that you have a compute node in multiple availability zones, and a storage layer that is shared across all of your different compute nodes. Within the storage layer, your data is replicated multiple times within an availability zone (AZ), and again across other availability zones. So you will have at least 6 copies of your data (2 in each AZ). If you really want to go down a rabbit hole on how this is done, it uses the quorum and gossip protocols.
