A blade server host a chasis of blades, computing, networking, storage, etc. These blades are managed as awhole usually. Does openstack have hirarchical management/operational scheme? Does it handle plugin such a scheme?
Hi,
The simple answer is that right now, OpenStack does not treat blades in any special way.
If the hypervisor in use is blade aware, this may make a difference, but in general each server appears as a server. It is possible to group servers together into HostAggregates (http://docs.openstack.org/trunk/openstack-compute/admin/content/host-aggregates.html) to take some different action scheduling-wise, or manage taking entire blade chassis offline at the same time.
It sounds like you have a good idea of the actions that can be performed with blade servers that OpenStack could be taking advantage of. It would be great to get you to share your ideas with the OpenStack developers to improve the current situation. The development mailing list is found here: https://wiki.openstack.org/wiki/MailingLists#Future_Development
Asked: 2013-05-22 12:42:26 -0600
Seen: 650 times
Last updated: May 30 '13
Further more, how does the openstack know if two blades are configured in a redundancy group (one is a hot standy of the other)?