Managed Databases
PostgreSQL, MySQL, and Mongodb managed for you. Backups, patches, and failover handled.
Pricing
Plans combine compute and additional disk space SKUs. Each database engine sizes compute and additional disk space independently so you grow CPU, memory, or disk at its own pace without repurchasing the whole plan.
Loading SKUs...
Ship schema changes,
not infrastructure tickets.
Running a production database means patch cycles, WAL management, replication lag monitoring, failover testing, and a rotation of operational tasks that rarely appear in a product roadmap. Managed Databases absorbs that surface.
You connect with a standard connection string. We handle the engine version, automated backups, point-in-time recovery, and failover. Schema migrations and query tuning stay your responsibility; everything below the connection string is ours.
PostgreSQL, MySQL, and MongoDB are available with the same sizing model as compute: you pick a CPU and memory SKU, attach additional disk space, and pay for what you provision.
EPYC compute, ECC memory,
NVMe WAL storage.
Database nodes run on AMD EPYC 4th Gen processors, matching the hardware class used across the Advance product line. DDR5 ECC on-die memory protects in-flight buffer pool state from correctable errors, which matters for PostgreSQL shared_buffers and MySQL InnoDB buffer pool sizing.
WAL segments and data directories write to NVMe SSD in soft RAID. Sequential write throughput is consistent and predictable, keeping replication lag tight under sustained ingest. Additional disk space scales independently from compute so you grow each dimension at its own pace.
Automated backups run daily to durable object storage with 30-day retention by default. Point-in-time recovery uses WAL archiving so you can restore to any second within the retention window.
What you get on Day One
Three database engines, automated operations, and hardware-isolated compute on the same EPYC stack.
Three engines
PostgreSQL (14+), MySQL (8.0+), and MongoDB (7.0+). Each on the same EPYC hardware baseline with independent compute and additional disk space SKUs.
Automated backups
Daily snapshots plus continuous WAL archiving for PostgreSQL and MySQL. Restore to any second within the 30-day retention window.
High availability
Multi-node deployment with synchronous replication and automatic failover. Promotes the replica on primary failure without manual intervention.
Encryption
AES-256 at rest on all volumes and backup storage. TLS required for all client connections. No plaintext paths in transit or at rest on the platform.
Configuration & Platform Details
A concise reference for architecture reviews, security questionnaires, and capacity planning.
Platform specifications
Values reflect the Managed Databases product line; exact limits may vary by region, engine, and account tier.
- Engines
- PostgreSQL 14+, MySQL 8.0+, MongoDB 7.0+. Engine version selectable at creation. Major version upgrades via API.
- Compute platform
- AMD EPYC 4th Gen. Dedicated vCPU and memory per instance. No shared resources with adjacent database instances on the same host.
- Memory
- DDR5 ECC on-die, matched to the selected SKU. Sized to buffer pool and sort memory requirements.
- Storage
- NVMe SSD in soft RAID for data directories and WAL. Additional disk space is sized independently from compute. Multi-drive redundancy for each volume.
- Backups
- Daily automated snapshots to durable object storage. Continuous WAL archiving for PostgreSQL and MySQL. Point-in-time recovery up to 30 days.
- High availability
- Synchronous streaming replication for PostgreSQL and MySQL. Automatic failover within the same region. Replica promotion via health check on primary.
- Encryption
- AES-256 encryption at rest for data directories and backup storage. TLS 1.2+ required for all client connections.