Hello, I want to ask if it is possible to create one large disk with several clusters that I have. I have two servers, both have 1TB ssd and I would like to connect one server to the other in order to increase the size of the disks, it is known that I would connect the second server to one using disks and get 2TB.
Combining multiple disks from other servers
- Thread starter damiko194
- Start date
The technology you are looking for is "distributed storage". But it will only get you there half way. To have reliable distributed storage the data need to be replicated and recoverable in case one server fails. You can look into Ceph, since its configuration comes built-into PVE.
To figure out how much space will be available after data replication you can use an online calculatore, ie https://florian.ca/ceph-calculator/
Note that Ceph was designed by large entities, for large number of nodes.
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
To figure out how much space will be available after data replication you can use an online calculatore, ie https://florian.ca/ceph-calculator/
Note that Ceph was designed by large entities, for large number of nodes.
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
yes, Ceph is an Open Source product, and is available without subscription within PVE.
Keep in mind that with 3 servers, each with 1TB disk, a properly configured Ceph setup will produce 1TB of total available space.
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
Yes, that is an artifact of data protection across networked devices.
There are hacks you can put in place, using duct tape and chewing gum. I.e. export each disk via NFS, mount them on single server/vm using subdirectories and pretend its one large space.
Even locally within a single server, if you have 2 disks which are 1TB each - there is no safe technology to present them as single 2TB space. You could use RAID0-like technology, which would mean a single disk failure will cause total data loss.
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox