Le 23/01/2014 18:05, Day, Phil a écrit :
I think there is a cross-over point here somewhere - call it "managed cloud" where the user doesn't want to run a cloud but does want (and is willing to pay for) some guarantee of isolation of hypervisors, but also wants to share other aspects of the cloud (images, volumes, VPC networks). They get a degree of controlled Admin (can set up their own flavors for example) - but they are still only doing configuration type tasks, not operating the system as such - and probably most significantly they can build systems that are a combination of instances running on shared and dedicated servers. That's kind of where pCLouds was evolving to - and I think you can build such a beast more effectively by running multiple dedicated Nova's configured to take to a shared Glance/Cinder/Neutron. I.e this belongs on-Nova, not in-Nova
Well, it occurs to me that the main benefic of having dedicated hosts for running your instances is that you can handle the load, and you're not dependent of others. That's particularly important when, as a company or as a SRE team, you want to exactly know your exact capacity in terms of requests per secs vs. CPU load.
Phil, I would be glad joining you for implementing this, would you mind if we discuss it directly ?
-Sylvain _______________________________________________ OpenStack-dev mailing list OpenStack-dev@lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev