1. Does anyone have i/o bench marks comparing network and disk throughput when using Linux versus NetBSD as Dom0? I don't have any real benchmarks, but with NetBSD as dom0 and domU I've observed < 10% loss in dd rate of xbd0d compared to dd from the file in the dom0, and another < 10% loss from dd from that file compared to dd From the raw disk. 2. We know we can use the xbd driver to provide virtual disks in NetBSD, but are other storage technologies mature in the NetBSD environment? I'm thinking aboutthings like ISCSI, NFS, or other network storage technologies. I use files in the domU filesystem (RAID1 raidframe on local sata), and that has been both satisfactory and solid. Using NFS seems scary (not about NetBSD - just putting NFS in the path from an OS to what it thinks is a disk). My understanding is that as long as you have block device (including vnd) to hand to xen, all is well. So vnd of a file in NFS should work, and vnd of a file in iscsi, or an iscsi block device should also work. I would suggest that setting it up to try with your storage setup should be pretty easy. One thing that has been trouble in the past and may still be trouble is vnd with sparse files. 0. [missing question about reliability] My group has had 3-4 xen boxes, each with NetBSD dom0, with the oldest going back to 2005. It's been highly reliable, with the dom0 typically being up for on the order of a year until we have grid power issues. There have been crashes, but less than 1/year/dom0.
Attachment:
pgpqtSIfMMVVF.pgp
Description: PGP signature