CPU and RAM Hot-Plugging

Registered by wlfightup

Now these operating systems support hot-swappable cpu and memory, and customers have a strong demand for these.

At the same time I think that the reserve resources to support the hot plug that is bad way, because it is a waste of physical resources, it is unreasonable, And directly let vm resize to the bigest What is the difference?. So reserved resources to vm expansion is very unreasonable.

I have two ideas:
1, api --> scheduler -- > if resource is insufficient --> prompts for hot migration --> then horizon calls hot-plug interfaces to implement live-resize
2, api -> scheduler -> resources enough -> call libvirt to resize

Blueprint information

Status:
Not started
Approver:
None
Priority:
Undefined
Drafter:
wlfightup
Direction:
Needs approval
Assignee:
wlfightup
Definition:
New
Series goal:
None
Implementation:
Unknown
Milestone target:
None

Related branches

Sprints

Whiteboard

(?)

Work Items

This blueprint contains Public information 
Everyone can see this information.

Subscribers

No subscribers.