Per aggregate scheduling weight

Registered by Zhenyu Zheng on 2018-09-03

Filter Scheduler uses the so-called weights during its work. A weigher is a way to select the best suitable host from a group of valid hosts by giving weights to all the hosts in the list.

In the current implementation, the weights are controlled by config options like
[filter_scheduler] cpu_weight_multiplier
the total weight of a compute node is calculated by combination of serveral weighers:
weight = w1_multiplier * norm(w1) + w2_multiplier * norm(w2) + ...

As it is controlled by config options, the weights are global across the whole deployment,
this is not convinient enough for operators and users.

This blueprint propose to allow set weights on host aggregates to have more flexibility
when scheduling.

Blueprint information

Status:
Complete
Approver:
Matt Riedemann
Priority:
Low
Drafter:
Zhenyu Zheng
Direction:
Approved
Assignee:
Yikun Jiang
Definition:
Approved
Series goal:
Accepted for stein
Implementation:
Implemented
Milestone target:
milestone icon stein-3
Started by
Matt Riedemann on 2019-01-03
Completed by
Matt Riedemann on 2019-01-24

Related branches

Sprints

Whiteboard

Gerrit topic: https://review.openstack.org/#q,topic:bp/per-aggregate-scheduling-weight,n,z

Addressed by: https://review.openstack.org/599308
    Per aggregate scheduling weight (spec)

Approved for Stein. -- mriedem 20181106

Addressed by: https://review.openstack.org/628163
    WIP: Per aggregate scheduling weight

(?)

Work Items

This blueprint contains Public information 
Everyone can see this information.

Subscribers

No subscribers.