Skip to main content
Change how much capacity your app has: more or fewer machines in the cluster, different machine types, or more copies of a service. Machines are added and removed only when you ask, review the plan and approve. A service can scale its own number of copies within limits you set in its template (see below). The orchestrator also restarts and reschedules managed workloads within its policies. See Watcher and recovery.

​
Before you start

​
Ask Monk

Add machines to the cluster:
Remove machines:
Change machine size. Add nodes of the new size, move work onto them, then remove the old ones:
Run more copies of a service or give it more resources. These are settings in the MANIFEST. Your coding agent edits it and Monk deploys the change:
Scale a service automatically. A scaling rule in the service’s template lets the orchestrator add or remove copies by CPU, memory or disk use, between a minimum and a maximum you choose. Copies spread across the cluster’s nodes, so a high maximum needs enough nodes to hold them:
Give a volume more space. Change the volume size in the template and deploy. Growing a cloud volume works on AWS and GCP:
Get a recommendation:
Check the valid regions, machine types and GPU types first if you’re not sure:

​
What you review

The plan in the local dashboard shows the machines to add or remove, the services that move, and the estimated cost after the change. Removing a node can move or stop the workloads on it, so check the list. Nothing changes until you approve.

​
What you get

The cluster and services at the new size. Check them:

​
Upgrading Monk on the cluster

Nodes are upgraded one at a time.

​
Next