In a set of servers (some brand new dell T440 with 64GB or RAM, 2 500M SSD ZFS Raid 1, three 4TB HDD in ZFS raid3) we are catching some performance issue, eg in some operation we catch 10/20 MB/s maximal performance, with load that go sky high (load 40).
An example: a couple of identical servers, connected via single gigabit connection. I've setup on server B a NFS server on HDD ZFS pool, then a NFS type storage (on both servers, they are clustered). I've copied from my PC a PVE VM backup, roughly 500GB of data, via NFS to server B: 80-90 MB/s, load 2-4. I've restored the VM on server A, thus *reading* data fron NFS, and restoring again on HDD ZFS pool on server A: 10-20 MB/s transfer rate, load at 40. This make no sense to me. As subject say, there's some 'Disk performance test guidance' document, so i can test and find the bottleneck? Thanks. -- Consolatevi! Sul sito http://www.sorryeverybody.com migliaia di americani chiedono scusa al mondo per la rielezione di Bush. (da Cacao Elefante) _______________________________________________ pve-user mailing list [email protected] https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-user
