Hey everyone,
I am looking to VM placement problems to solve, and i would like to know how this problem is considered in Openstack community, and how it is solved ?
If my question is not obvious, i can give more details.
Thanks in advance,
Sincerely
Look at Host Aggregates, Availability zones and SchedulerFilters features in Openstack
http://docs.openstack.org/trunk/openstack-compute/admin/content/scheduler-filters.html
http://docs.openstack.org/trunk/openstack-ops/content/availability_zones.html
http://docs.openstack.org/trunk/openstack-compute/admin/content/host-aggregates.html
Ok , thanks a lot. To be more precisely on my question, i would like to know how the RAM performance is monitored to the placement solution ? And, why the openstack placement solution is not proceeding as it does by Best first decreasing, or first fit increasing methods ? Thanks for your answers.
Please refer to this DeveloperWorks entry: Nova Schedule and its algorithm for some detail related to nova scheduler.
Ok , thanks a lot. To be more precisely on my question, i would like to know how the RAM performance is monitored to the placement solution ? And, why the openstack placement solution is not proceeding as it does by Best first decreasing, or first fit increasing methods ? Thanks for your answers.
Marc
Asked: 2013-05-31 05:13:57 -0500
Seen: 292 times
Last updated: Jun 12 '13
Hi , it would be great if you could provide more details.
If i have well understand the placement solution provided by Openstak, yhe VM placement consists to randomly assign a VM to an available host.
Thus, i would like to know how the different measures or performances (RAM, CPU, Storage, etc) of each server in an infrastructure are monitored and taken into account to place correctely the arrived VMs in the right server ?
@Marc if you're the one that asked the original question, please consider updating the question instead of adding details in the comments: this makes things more readable.