High Availability with MariaDB

Clustering, replication and automatic failover

MariaDB Enterprise Server is the first and only database (enterprise open source or proprietary) with advanced high availability features comparable to Oracle Database Enterprise Edition, and capable of supporting the development and operation of mission-critical applications. MariaDB SkySQL cloud database combines Kubernetes and MariaDB MaxScale to maintain availability across instance/zone failures and ensure all database deployments are restored to full capacity, regardless of whether the failures affected primaries or replicas.

The definitive guide to high availability with MariaDB Enterprise Server

This white paper provides a detailed and comprehensive overview of the enterprise high availability (HA) features in MariaDB Enterprise Server, and explains the trade-offs and considerations when choosing an HA strategy.

Read more

Why choose MariaDB Enterprise Server when high availability is required

Clustering and replication

MariaDB Enterprise Server has multi-master clustering similar to Oracle RAC (but without the need for an expensive SAN, a single point of failure) as well as semi-synchronous replication and active standby databases like Oracle Active Data Guard or Microsoft SQL Server Availability Groups.

Automatic failover

MariaDB Enterprise Server supports automatic failover with its advanced database proxy, MariaDB MaxScale. If the primary database fails or is otherwise unavailable/unreachable, MariaDB MaxScale will promote one of the secondary databases and begin routing queries to it.

Connection and session failover

MariaDB Enterprise Server goes beyond automatic failover, with MariaDB MaxScale, its advanced database proxy, capable of failing over both database connections and active sessions to ensure mission-critical applications are not interrupted if an automatic failover occurs.

Transaction replay

MariaDB Enterprise Server, like Oracle Database with Oracle Application Continuity, can restore in-flight transactions after an automatic failover using MariaDB MaxScale, its advanced database proxy, so applications do not have to restart incomplete transactions from the beginning.

Transparent topology changes

MariaDB Enterprise Server hides all topologies changes from applications so developers do not have to write code to handle database and/or server failures. Applications are neither aware of nor interrupted by database or server failures, everything is handled by MariaDB MaxScale.

Denial of service protection

MariaDB Enterprise Server not only ensures availability in the event of a database or server failure, it can prevent denial of service attacks by blocking or stopping malicious queries intended to reduce database availability with MariaDB MaxScale, its advanced database proxy.

High availability with multi-master clustering and adaptive routing

In the example topology below, MariaDB Enterprise Server is deployed as a multi-master cluster with three database instances/nodes and uses read/write splitting to scale reads by routing writes to the primary database and load balancing reads across the secondary databases.


MariaDB Maxscale High Availability Topology

How MariaDB Enterprise Server meets high availability requirements

In this webinar, we'll provide an in-depth overview of the high availability features in MariaDB Enterprise Server and a first look at the advanced options introduced in MariaDB Enterprise Server (e.g., transaction replay).

Watch Now
Download Contact