Thanks for that. I think we will leave this off for now to do further testing as I suspect the penalty might not be worth it. The hardware is just before AES-NI was introduced so the encryption would be in software.
- Pieter
On 3 March 2014 11:57, Mariusz Gronczewski <mariusz.gronczewski@xxxxxxxxxxxxx> wrote:
Dnia 2014-03-03, o godz. 10:41:04
Pieter Koorts <pieter@xxxxxxxxxxxx> napisał(a):
performance will depend on use case and if your CPU have AES-NI and
> Hi
>
> Does the disk encryption have a major impact on performance for a
> busy(ish) cluster?
>
> What are the thoughts of having the encryption enabled for all disks
> by default?
>
> - Pieter
your kernel, cryptsetup benchmark should show something like this:
# Algorithm | Key | Encryption | Decryption
aes-cbc 128b 586,0 MiB/s 2114,6 MiB/s
serpent-cbc 128b 87,3 MiB/s 288,8 MiB/s
twofish-cbc 128b 180,0 MiB/s 343,1 MiB/s
aes-cbc 256b 428,0 MiB/s 1601,0 MiB/s
serpent-cbc 256b 93,9 MiB/s 296,3 MiB/s
twofish-cbc 256b 189,2 MiB/s 351,6 MiB/s
aes-xts 256b 1758,8 MiB/s 1791,3 MiB/s
serpent-xts 256b 309,0 MiB/s 294,9 MiB/s
twofish-xts 256b 345,7 MiB/s 346,7 MiB/s
aes-xts 512b 1406,2 MiB/s 1411,7 MiB/s
serpent-xts 512b 313,7 MiB/s 295,6 MiB/s
twofish-xts 512b 347,8 MiB/s 350,2 MiB/s
so if you have a lot of OSDs per machine and do a lot of sequential IO
you might hit into CPU wall in performance
--
Mariusz Gronczewski, Administrator
efigence S. A.
ul. Wołoska 9a, 02-583 Warszawa
T: [+48] 22 380 13 13
F: [+48] 22 380 13 14
E: mariusz.gronczewski@xxxxxxxxxxxx
<mailto:mariusz.gronczewski@xxxxxxxxxxxx>
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
