Choosing your cluster type
Suggest editsEDB Postgres AI Cloud Service supports different cluster types and configurations.
Cluster types
Expand each cluster type to obtain more information.
Single instance
Single instance databases are great for development and testing, but for production workloads, you need to consider high availability and fault tolerance.Primary/secondary replication
Primary/secondary replication is a common high-availability solution for databases. In this configuration, a primary database server is responsible for processing read and write requests. A secondary database server is setup to replicate the primary database server. If the primary database server fails, the secondary database server can take over and become the primary database server.This configuration provides fault tolerance and high-availability in a particular location. This is available with EDB Postgres® Advanced Server (EPAS) and EDB Postgres Extended Server (PGE).
Distributed high availability
High availability is a critical requirement for mission-critical workloads. EDB Postgres Distributed (PGD) provides a distributed database environment designed to ensure high availability and fault tolerance for mission-critical workloads. PGD can use EPAS, PGE, or PostgreSQL databases as the underlying replicated database. PGD is available for self-managed deployment and on the EDB Postgres AI Cloud Service (as the Distributed High Availability option).Cluster types and configurations
Choose the cluster type and configuration that meets your availability requirements.
Select the cluster type names in the table heading for more information on each configuration.
| Consideration | Single node | Primary/Standby HA | HA + standby replica | Distributed HA single region | Distributed HA multi-region |
|---|---|---|---|---|---|
| Data replication | None | Physical | Physical | Logical | Logical |
| Region | Single | Single | Multi | Single | Multi |
| VM failure tolerance | ✅ | ✅ | ✅ | ✅ | ✅ |
| AZ failure tolerance | TBD | ✅ | ✅ | ✅ | ✅ |
| Region failure tolerance | TBD | TBD | ✅ | TBD | ✅ |
| Recovery time objective | varies | 35s-60s | varies | 0 | 0 |
| Recovery point objective | <5 min | 0 | <5 min | 0 | 30s (configurable) |
| Service level agreement | 99.5% | 99.99% | 99.99% | 99.99% | 99.995% |
- On this page
- Cluster types
- Cluster types and configurations
Could this page be better? Report a problem or suggest an addition!