Scale

Dynamically handle unplanned workload changes

Key Features

One operational console for your MySQL, MariaDB, PostgreSQL & MongoDB databases

Elasticity

Easily add or remove a node with a single action

Workload Distribution

Add slaves to offload your databases, or use as a backup node

Clone Databases

Create copies of your entire cluster that you can transition from dev to QA and on to production

Scale

Elasticity

Handle unplanned workload changes by dynamically scaling out with more nodes. Optimize resource usage by scaling back nodes.

Let us worry about the mechanics to get it done ,including reconfiguring load balancer configurations to match the topology changes.

Scale

Workload Distribution

Add a slave to your cluster with the master dataset automatically primed, and asynchronous replication running. Use a delayed slave as a live backup node, or as a loosely coupled reporting server that does not impact cluster performance.

Scale

Clone Databases

Efficiently duplicate your production database to support application development and testing activities with the latest dataset.

Other scenarios include live migrations to new hardware or datacenters, troubleshooting of production datasets in the safety of a test environment or upgrade/patch test of new database software versions.

Frequently Asked Questions

When adding nodes to a cluster, do I need to install the database software on the new hosts?

No, all you need are clean VMs with the OS installed. ClusterControl will connect to the host, install the necessary software and make sure it synchronises with the rest of the cluster.

Are load balancers kept updated when adding and removing nodes?

If the load balancers are managed by ClusterControl, then yes. The list of database nodes is maintained to reflect any topology changes. If you deployed the load balancer using ClusterControl, it also has health checks customised for the cluster type.

Why would I want to add an asynchronous slave to a Galera Cluster?

There are a few reasons. Long-running reporting/OLAP type queries on a Galera node might slow down an entire cluster, if the reporting load is so intensive that the node has to spend considerable effort coping with it. So reporting queries can be sent to a standalone server, effectively isolating Galera from the reporting load. In a belts and suspenders approach, an asynchronous slave can also serve as a remote live backup.

Do you have an API to handle dynamic scaling of the database?

Yes, you can easily create an autoscaling advisor to monitor your custom performance metrics and make scaling decisions based on that. One example of how to scale a replication setup can be found here.