> We are at the end of the process of designing and purchasing storage to > provide Ceph based backend for VM images, VM boot (ephemeral) disks, > persistent volumes (and possibly object storage) for our future Openstack > cloud. > We considered many options and we chose to prefer commodity storage server > vendors over ‘brand vendors’. There are 3 (+1 extra) types of storage server > options we consider at the moment: Just some general recommendations: * Make sure the hardware matches the expected workload: E.G. running a database or mailserver with continuous IO is totally different from a webserver which only does a few reads when it boots. Having lots space but slow spinning big disks (4TB) might not be the best solution for all scenarios
* BIG nodes have big impact when they fail: Rebalancing because of a lost node with 120TB of storage will take a LOT of time. In general I would prefer more smaller nodes then a few big ones because of this. Cheers, Robert _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com