Bug 1041368

Summary: [RFE][nova]: Add ecu support for nova based on Cgroup on kvm hypervisor
Product: Red Hat OpenStack Reporter: RHOS Integration <rhos-integ>
Component: RFEsAssignee: RHOS Maint <rhos-maint>
Status: CLOSED UPSTREAM QA Contact:
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: unspecifiedCC: markmc, yeylon
Target Milestone: ---Keywords: FutureFeature
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
URL: https://blueprints.launchpad.net/nova/+spec/nova-ecu-support
Whiteboard: upstream_milestone_none upstream_status_unknown upstream_definition_obsolete
Fixed In Version: Doc Type: Enhancement
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2015-03-19 17:31:33 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description RHOS Integration 2013-12-12 15:37:51 UTC 
Cloned from launchpad blueprint https://blueprints.launchpad.net/nova/+spec/nova-ecu-support.

Description:

Currently nova supports cpu qos with vcpu pin and qouta/period, it is not so intelligent, and the cloud user usually doesn't know how to set the 'quota/period' of all instances and could still keeps fair and performance of them.
We should add an intelligent(non or very few manually interactive), fair, consistency performance way to implement the ecu concept like amazon ec2, we also need to treat ecu as an extra_spec of flavor, the instances which has the same ecus should have consistency performance even if they are running on several heterogeneous servers.
We also need to treat ecu as a resource like vcpu, it need has qouta management, resource tracking, scheduler filter.

Specification URL (additional information):

None