Amazon Aurora - The Relational Database optimized for the Cloud
Amazon Aurora is a relational database, compatible with MySQL and PostgreSQL and created by Amazon for the Cloud. Amazon Aurora is up to 5x faster than a standard MySQL database and 3x faster than a standard PostgreSQL database.
It relies on Amazon RDS and extends its features.
Features of Amazon Aurora
High performance and scalability
- High Performance:
- It is connected to the Cloud
- It supports Low Latency reading with up to 15 Read Replicas in Multi-AZ
- Depending on the option chosen, one can have 1 Master in Writing or several connected to the same storage volume
- Scaling:
- Automatic Storage Auto Scaling increase in 10 GB increments up to 128 TB
- The number of Read Replicas can be managed by an Auto Scaling Group
- A Serverless option allows you to manage intermittent and unpredictable workloads
Availability
- High Availability:
- 6x data replication in a shared storage volume on 3 AZs
- Continuous backup in Amazon S3
- 1 Endpoint for Writing and 1 for Reading
- Automatic failover in case of unavailability of an AZ in less than 30s
- Support the cross-Region (Database Location: Global option);
- Data are replicated across multiple Regions
- Write operations on the Primary Region are replicated in the Secondary Regions with latency < 1s
- Automatic failover in case of unavailability of a Region
- Data can be restored at any point of time whatever the time of the last backup (Backtrack option)
Security
- Seamless maintenance of Aurora instances
- Network isolation with Amazon VPC
- Data encryption at rest and in transit with AWS KMS as well as storage volumes, backups, snapshots and replicas
