qinling

Autoscaling support for function service

Registered by Lingxian Kong

Now there is only one-to-one mapping from function to pod in k8s as default orchestrator, it can not scale well if there are a lot of invokes to the function.

Blueprint information

Status:
Complete
Approver:
Lingxian Kong
Priority:
High
Drafter:
Lingxian Kong
Direction:
Approved
Assignee:
Lingxian Kong
Definition:
Approved
Series goal:
Accepted for queen
Implementation:
Implemented
Milestone target:
milestone icon queen-3
Started by
Lingxian Kong
Completed by
Lingxian Kong

Related branches

Sprints

Whiteboard

Gerrit topic: https://review.openstack.org/#q,topic:bp/qingling-autoscaling,n,z

Addressed by: https://review.openstack.org/498309
    Add scale up ability for function execution

Addressed by: https://review.openstack.org/498971
    Add scale up ability for function execution

Addressed by: https://review.openstack.org/499976
    Scale down function

Addressed by: https://review.openstack.org/526330
    Function execution concurrency improvement

(?)

Work Items

Nearby

This blueprint contains Public information 
Everyone can see this information.