http://geekyschmidt.com/2011/01/09/openbsd-drive-encryption-benchmarks
Note that this is done using bonnie. Bonnie isn't very good in figuring out what I real world load looks like. It does give some insight in cpu usage. Crypto is slow, end of story. If you want encrypted disks you'll pay a performance price. Either your data is worth it or not. The thing to look for is if the performance degradation is acceptable. I don't notice any speed decrease using softraid day to day which is writing code and surfing the web. Your load is likely different and have different characteristics. On Sat, Mar 19, 2011 at 02:49:52PM +0000, Kevin Chadwick wrote: > On Fri, 18 Mar 2011 16:58:59 +0000 > Kevin Chadwick wrote: > > > I do get a fair increase in cpu usage for a disk at full speed disk with > > vnd but it's acceptable. Have people already done cpu usage and > > transfer speed comparisons to save me further tests. > > Well I was about to run a comparison test on vmware and I'm well > confused unless it's a strange vmware bug or maybe the dynamic size disk > mechanism. I might have to pull out a box. > > > SO I took a free partiton at the end of the disk split it in two > (approx) one for svnd and one for bioctl. > > I wiped them but got a speed difference which obviously scuppers any > potential tests. Tried reversing the orders in case the hard drive was > ready the second time but get consistent but unexpected results. > > # /bin/dd if=/dev/zero of=/dev/rwd0m bs=1m > dd: /dev/rwd0m: short write on character device > dd: /dev/rwd0m: end of device > 397+0 records in > 396+1 records out > 415334400 bytes transferred in 4.604 secs (90192757 bytes/sec) > > # /bin/dd if=/dev/zero of=/dev/rwd0n bs=1m > dd: /dev/rwd0n: short write on character device > dd: /dev/rwd0n: end of device > 391+0 records in > 390+1 records out > 409600000 bytes transferred in 2.574 secs (159069629 bytes/sec) > > > now wd0n first gives the same results > > > # /bin/dd if=/dev/zero of=/dev/rwd0n bs=1m > dd: /dev/rwd0n: short write on character device > dd: /dev/rwd0n: end of device > 391+0 records in > 390+1 records out > 409600000 bytes transferred in 2.545 secs (160931390 bytes/sec) > > # /bin/dd if=/dev/zero of=/dev/rwd0m bs=1m > dd: /dev/rwd0m: short write on character device > dd: /dev/rwd0m: end of device > 397+0 records in > 396+1 records out > 415334400 bytes transferred in 4.339 secs (95713197 bytes/sec) > > I get the same with write caching off and wd0n shortened to not be > allocated to the very end of the disk with the same results. > > > Anyone have any ideas or compared bioctl and svnd speed and cpu usage in > the past? > > I read svnd may be marginally quicker but I would expect bioctl to use > less cpu and be quicker?