> For the complete documentation index, see [llms.txt](https://docs.platform9.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.platform9.com/private-cloud-director/2026.4/virtualized-clusters/virtualmachine/attaching-virtual-persistent-memory-to-guests.md).

# Attaching Virtual Persistent Memory To Guests

The virtual persistent memory (vPMEM) feature in <code class="expression">space.vars.product\_name</code> enables administrators to configure vPMEMs for virtual machines using physical persistent memory (PMEM) that can provide virtual devices.

Note that the vPMEM feature is not enabled by default in your <code class="expression">space.vars.product\_name</code>. Contact your <code class="expression">space.vars.product\_name</code> support liaison if you would like this feature enabled.

### Pre-requisites

* Persistent Memory Hardware
* Following modules loaded at the linux kernel level for hypervisor hosts:

  `dax_pmem`, `nd_pmem`, `device_dax`, `nd_btt`

### Configure PMEM Namespaces

Your first step is to configure your PMEM namespaces as vPMEM backends using the `ndctl` utility.

```bash
$ sudo ndctl create-namespace -s 30G -m devdax -M mem -n ns3
{
  "dev":"namespace1.0",
  "mode":"devdax",
  "map":"mem",
  "size":"30.00 GiB (32.21 GB)",
  "uuid":"937e9269-512b-4f65-9ac6-b74b61075c11",
  "raw_uuid":"17760832-a062-4aef-9d3b-95ea32038066",
  "daxregion":{
    "id":1,
    "size":"30.00 GiB (32.21 GB)",
    "align":2097152,
    "devices":[
    {
      "chardev":"dax1.0",
      "size":"30.00 GiB (32.21 GB)"
    }
    ]
  },
  "name":"ns3",
  "numa_node":1
}
```

Now you can list the available PMEM namespaces on the host:

<pre class="language-bash"><code class="lang-bash"><strong>$ ndctl list -X
</strong>[
  {
    ...
    "size":6440353792,
    ...
    "name":"ns0",
    ...
  },
  {
    ...
    "size":6440353792,
    ...
    "name":"ns1",
    ...
  },
  {
    ...
    "size":6440353792,
    ...
    "name":"ns2",
    ...
  },
  {
    ...
    "size":32210157568,
    ...
    "name":"ns3",
    ...
  }
]
</code></pre>

### Configure Memory Resource Classes

Now that you have the namespaces created, the next step is to specify which PMEM namespaces should be available to virtual machines. You can do this by defining one or more PMEM resource classes in the compute service advance configuration [as specified here](/private-cloud-director/2026.4/virtualized-clusters/nova-override.md#configure-pmem-namespace).

Make sure to restart the compute service on each host after doing the above configuration.

### Configure a flavor

Now that you have the required PMEM namespaces and classes configured, you can specify a comma-separated list of the `$LABEL`s that correspond to the resource classes you defined in the step above, to the flavor’s `hw:pmem` property. Multiple instances of the same label are permitted.

```bash
pcdctl flavor set --property hw:pmem='6GB' my_flavor
pcdctl flavor set --property hw:pmem='6GB,LARGE' my_flavor_large
pcdctl flavor set --property hw:pmem='6GB,6GB' m1.medium
```

Based on the above examples, a `pcdctl server create` request with `my_flavor_large` will spawn a new virtual machine with two vPMEMs. One, corresponding to the `LARGE` label, will be `ns3`; the other, corresponding to the `6G` label, will be arbitrarily chosen from `ns0`, `ns1`, or `ns2`.

\ <br>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.platform9.com/private-cloud-director/2026.4/virtualized-clusters/virtualmachine/attaching-virtual-persistent-memory-to-guests.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.