Thanks for the link !
some amazing benches around :) Let's hope we'll have more details on the methodology, but some results are stunning

Razique Mahroua - Nuage & Co
Tel : +33 9 72 37 94 15


Le 17 janv. 2013 à 01:54, Blair Bethwaite <blair.bethwa...@gmail.com> a écrit :

On 17 January 2013 09:48, Sean Bigdatafun <sean.bigdata...@gmail.com> wrote:
IMO, it's always a tradeoff. I am just very curious how Amazon configures its EC2 hardware. 

Good luck getting that information! You might be able to make some guesses from benchmark info though, there is plenty around. This post from Scalyr takes a great all round look at EC2 IO performance, and it's pretty recent data: http://blog.scalyr.com/2012/10/16/a-systematic-look-at-ec2-io/ .

Personally, I would guess that for ephemeral drives they use either no RAID at all or RAID10, probably none though, especially as they don't advertise any reliability for these drives. The issue of rebuild performance in parity-based RAID configs and the effect it would have on guest performance would be a significant consideration. Having said that, it's difficult to find reports of ephemeral drive failures, and Amazon don't quote any failure rates themselves.

--
Cheers,
~Blairo
_______________________________________________
Mailing list: https://launchpad.net/~openstack
Post to     : openstack@lists.launchpad.net
Unsubscribe : https://launchpad.net/~openstack
More help   : https://help.launchpad.net/ListHelp

_______________________________________________
Mailing list: https://launchpad.net/~openstack
Post to     : openstack@lists.launchpad.net
Unsubscribe : https://launchpad.net/~openstack
More help   : https://help.launchpad.net/ListHelp

Reply via email to