[PVE-User] (Very) basic question regarding PVE Ceph integration

Yannis Milios yannis.milios at gmail.com
Sun Dec 16 15:41:48 CET 2018

That's up to you to decide, PVE supports both hyper-converged setups (where
compute and storage nodes share the same hardware) and scenarios where
compute/storage nodes are separate.
You can choose for example to have 3 nodes in a PVE cluster, acting as
compute nodes and 3 separate nodes for the Ceph cluster, acting as storage
The main difference is that the costs involved in this scenario are much
higher compared to the hyper-converged setup.
It's also possible to manage Ceph clusters by using PVE gui (i.e use the
webgui for the Ceph tasks only).

Clearly in a hyper-converged setup, as the host(s) resources will be shared
between the compute and storage layer, proper measures in the design and
capacity planning must be taken in order to avoid downgraded performance.
For example ZFS is well known to consume high amounts of RAM.In a
hyper-converged setup, if improperly configured, it may consume half of
host RAM, potentially leaving VMs under pressure ..


On Sun, 16 Dec 2018 at 13:28, Frank Thommen <f.thommen at dkfz-heidelberg.de>

> Hi,
> I understand that with the new PVE release PVE hosts (hypervisors) can
> be used as Ceph servers.  But it's not clear to me if (or when) that
> makes sense.  Do I really want to have Ceph MDS/OSD on the same hardware
> as my hypervisors?  Doesn't that a) accumulate multiple POFs on the same
> hardware and b) occupy computing resources (CPU, RAM), that I'd rather
> use for my VMs and containers?  Wouldn't I rather want to have a
> separate Ceph cluster?
> Or didn't I get the point of the Ceph integration?
> Cheers
> Frank
> _______________________________________________
> pve-user mailing list
> pve-user at pve.proxmox.com
> https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user

More information about the pve-user mailing list