-
Feature
-
Resolution: Done
-
Critical
-
None
-
BU Product Work
-
False
-
False
-
0% To Do, 0% In Progress, 100% Done
-
0
Why is this important?
There are use cases out that require scaling based on metrics that are not either CPU and memory. For example, a database might be scaled based on the number of tables. Customers need more options when it comes to their own, or 3rd party, applications to configure an Horizontal Pod Autoscaler (HPA) object. That way, they can build high uptime and reliability SLOs using every tool available.
An example is scaling an app based on the number of HTTP connections.
Requirements
- Provide interface that allows admins to configure what metrics should be exposed via the custom metrics API.
- Exposed metrics should be available for use inside an HPA object.
Use case
https://docs.google.com/document/d/1PwgItr0Yqo5PqNCi5BhAf9q2ERwQXWzYUJeQtIEblNA/edit#
- incorporates
-
RFE-2116 Allow autoscaling via custom metrics
- Accepted
- is related to
-
OCPNODE-708 Tech preview of KEDA
- Closed
-
OCPNODE-705 Explore and Spike on KEDA
- Closed
- relates to
-
RFE-2116 Allow autoscaling via custom metrics
- Accepted
-
OBSDA-1 Allow autoscaling via custom metrics
- Closed
- links to