Manual and auto-scaling of VNF
On the event of failure or performance bottle necks, the deployed VNF needs to scale in/out to maintain the performance/
User would want to scale in/out based on their needs and manual scaling would help here.
Tacker does support monitoring the deployed VNF, and monitoring driver could be configured to automatically scale in/out the running VNF based on the identified events such as performance bottle-neck.
This spec is created to provide these feature to access using tosca defined. Also it would provide the require REST API for scaling.
Blueprint information
- Status:
- Complete
- Approver:
- Sridhar Ramaswamy
- Priority:
- High
- Drafter:
- Kanagaraj Manickam
- Direction:
- Approved
- Assignee:
- Kanagaraj Manickam
- Definition:
- Approved
- Series goal:
- Accepted for newton
- Implementation:
-
Implemented
- Milestone target:
-
newton-2
- Started by
- Kanagaraj Manickam
- Completed by
- Sridhar Ramaswamy
Related branches
Related bugs
Sprints
Whiteboard
Gerrit topic: https:/
Addressed by: https:/
VNF manual and auto scaling
Addressed by: https:/
Devref for vnf scaling feature
Addressed by: https:/
Adds limitations to scaling dev ref
Addressed by: https:/
Adds release notes for VNF scaling feature
Work Items
Work items:
Spec: https:/
Server - TOSCA definitions for scaling - https:/
Server - Refactor heat driver create() handler - https:/
Server - Heat driver config description updates - https:/
Server - REST API - https:/
Server - Heat driver update for scaling - https:/
Client - tackerclient scale CLIs - https:/
Doc - user guide : https:/
Server - Functional tests - https:/
Server - Unit tests - https:/
Server - Granular target VDU support : POSTPONED
Server - invoke mgmt-driver after scale out : POSTPONED
