Vultr Open Cluster Manager, designed to accelerate Vultr Cloud GPU cluster deployment, is now available as a one-click-deploy app on the Vultr Marketplace.

Powered by open source technologies including Terraform, Ansible, Grafana, and Slurm, Vultr Open Cluster Manager simplifies the process for getting AI workloads running on Vultr Cloud GPU clusters.

Getting started is a quick and easy process:

1. First, deploy the Vultr Open Cluster Manager marketplace app.

2. Next, inspect and modify the /root/config.yml file to specify the GPU plan, type, quantity, operating system, and data center region.

3. Then, initialize Terraform to build the cluster nodes, before running the Ansible playbook. This playbook updates the package repository on all cluster nodes and the cluster manager, installs and configures Grafana Alloy and the Slurm Daemon and Slurm Controller, and installs Prometheus Node Exporter on all nodes (with a dashboard installed on the local Grafana instance).

4. When that is completed, the cluster is ready for AI workloads.

For more details, see the Vultr Open Cluster Manager Tech Talk or the Vultr Open Cluster Manager Documentation.

Vultr is committed to simplifying infrastructure for AI and other applications. With Vultr Open Cluster Manager, available with Vultr’s one-click deploy feature in the Vultr Marketplace, deploying an AI-ready cluster on Vultr Cloud GPU is streamlined and accelerated.

To expedite AI-ready GPU cluster deployment, get started with Vultr Open Cluster Manager on the Vultr Marketplace today!