What Is Proxmox Datacenter Manager (PDM)?

Proxmox Datacenter Manager (PDM) is a central management tool that lets you monitor and partly control any number of independent Proxmox VE nodes and clusters from a single web interface. PDM aggregates the resources, metrics, and health status of the connected systems, offers a global search, and supports features such as live migration of virtual machines between separate clusters. It fills the gap between a single cluster and a site-spanning overview, playing a role similar to what VMware vCenter does in the VMware world.

One point up front, because many guides online still describe a pre-release state: PDM is no longer in alpha or beta. The first alpha appeared on December 19, 2024, beta 0.9 on September 11, 2025, and on December 4, 2025 Proxmox shipped the stable 1.0 release. Version 1.1 followed on May 28, 2026 (documentation at 1.1.4), adding central Ceph monitoring and a cross-remote guest overview, among other things. PDM is therefore usable in production, though as a young product it is not yet as broad in scope as tooling that has matured over many years.

PDM and a Proxmox cluster are not the same thing

The most common misconception is that PDM is a kind of super-cluster. It is not. A classic Proxmox cluster is a tightly coupled group of nodes. They share a common configuration through Corosync (pmxcfs), require quorum (a majority of votes), and need low network latency, which is why a cluster usually runs at a single site. Within that group you also get high availability with automatic failover.

PDM works on a completely different level. It couples multiple independent nodes or clusters only loosely through their APIs. Every connected system, called a "remote" in PDM terminology, stays fully autonomous and keeps its own quorum and its own HA. PDM is the cross-cutting control and visibility layer, not the cluster itself.

The rule of thumb: multiple nodes at the same site that should form HA together belong in a cluster. Multiple separate clusters or standalone nodes that you want to see and operate from one place are connected under PDM.

What Proxmox Datacenter Manager is for

PDM targets anyone running more than one Proxmox cluster or distributed standalone nodes. The most important capabilities in the 1.x line are:

• Central overview and metrics: A consolidated dashboard shows CPU, RAM, and storage I/O across all remotes in real time. Instead of logging into each individual web UI, you see utilization and health of the entire infrastructure in one place.
• Cross-cluster live migration: PDM can move running VMs not only within a cluster but also between different remotes without downtime. This is useful for maintenance windows, load balancing, or relocating workloads between sites.
• Role-based views and RBAC: Granular permissions and customizable views let you show each team only the resources relevant to them.
• Global search: A search function with a query syntax inspired by Elasticsearch and GitHub finds guests, nodes, and resources across the whole fleet.
• Centralized update management: PDM surfaces available patches across every connected system.
• Central SDN and EVPN configuration, along with basic lifecycle management (power actions on nodes and guests) and integration of Proxmox Backup Server instances.

Architecture: remotes, API tokens, and no single point of failure

PDM runs as its own small appliance, most easily installed from the official ISO, which ships a complete Debian system. The current 1.1 release is based on Debian 13 "Trixie" (13.5) with kernel 7.0 and is built in Rust with a Yew-based UI. For production use, Proxmox recommends at least 4 GiB of RAM and 40 GB of disk or more; for pure evaluation, 1 GiB of RAM and a little over 10 GB are enough. The requirement grows with the number of managed remotes and resources.

Connecting a remote happens through its API: you enter the URL of one node plus credentials or an existing API token, and PDM then automatically creates its own token for ongoing communication. PDM must be able to reach every remote directly over the network to send API requests and query metrics.

This loose coupling produces an important benefit: PDM is not a single point of failure for the managed infrastructure. If the PDM instance goes down, all clusters and nodes keep running fully autonomously, including their own high availability. What you lose during that time is only the central overview and the cross-cluster actions, not the operation of the VMs themselves.

Maturity and who benefits from PDM

PDM is free software under the GNU AGPLv3 and usable without license fees. For production use Proxmox recommends an Enterprise subscription, which provides access to the stable Enterprise repository and certified support. If you already have Enterprise support for your Proxmox remotes, you receive PDM updates and support under the same arrangement.

To be honest about its state: with 1.0, PDM left the pre-release phase and is stable, but it is a young product. Capabilities such as automatic load balancing across clusters (comparable to DRS in VMware) or full feature depth for every detailed task are not yet at the maturity that long-established tools offer. For many everyday tasks you still log into the individual cluster UI.

PDM makes sense for operators with multiple clusters or multiple sites, for MSPs, and for environments where a consolidated view and cross-cluster migration add real value. For a single cluster at one site, PDM brings little benefit, and the standard Proxmox VE interface is enough.

Operations and support

Whether an additional PDM makes sense in your environment, or whether a well-sized cluster already covers your requirements, depends on your sites, availability targets, and team structure. WZ-IT runs Proxmox in production customer environments and plans, builds, and maintains both clusters and multi-cluster setups. We deploy PDM, connect your remotes securely through API tokens, and define roles and views that match your teams. Learn more on our Proxmox services page or book a free initial consultation.