ZFS or BTRFS for Proxmox 3.1 ?

[jinjer]

Interesting readings which may help decide whether separate l2arc or zil will help you:
http://www.cuddletech.com/blog/pivot/entry.php?id=979
http://www.richardelling.com/Home/scripts-and-programs-1/zilstat

Yep... I use zilstat and arc_summary on my 'solaris' servers. Unfortunately you can't run them on linux, so at least I'm left guessing the correct sizes from "experience", but being my first linux zfs install, I'm not sure how much of the solaris stuff applies to linux.

 

[tarax]

CEPH is pretty much out of question since you are shooting for only one node setup as far as i can tell. In order to setup CEPH you need more than one node or else it is just plain waste of time with hardly any benefit.

That being said, can Ceph be setup and work reliably on one node ?
Because in this case the config would a t least be ready for scaling and HA.

Bests

 

[wahmed]

That being said, can Ceph be setup and work reliably on one node ?
Because in this case the config would a t least be ready for scaling and HA.

Bests

CEPH is really one node setup. At least the storage node portion is. You can use existing proxmox cluster to create Admin, MON nodes in VM format to used with CEPH. So you are not sacrficing machine for it. I personally use this setup. Only machine you need is for the CEPH OSDs themselves which are just hdds.
As for realiability, it really depends on how you define reliabilityin your situation. CEPH is 100% distributed storage system. It spreads data over multiple nodes and hdds to create redundancy. In 2 node setup, when one of the node goes down completely including all hdd in it, your storage will keep going due to its copy in one of the node. With one node setup it is all in one machine. Also CEPH works faster when you have multiple nodes with multiple hdds. The beauty of CEPH is, even if you start your setup with just one node, and it is time to expand, all you have to do is power up the new node, and from Admn console you just add new OSDs(hdds) into the existing cluster. ZERO Downtime. CEPH increases oveall storage spaces based on how many osds are connected. so your overall storage space is the combined space of all hdds in your cluster. Although some space is used to create replicas, but this is the basic idea.