Cloud Management and AIOps
This is an IBM Automation portal for Cloud Management and AIOps products. To view all of your ideas submitted to IBM, create and manage groups of Ideas, or create an idea explicitly set to be either visible by all (public) or visible only to you and IBM (private), use the IBM Unified Ideas Portal (https://ideas.ibm.com).
Shape the future of IBM!
We invite you to shape the future of IBM, including product roadmaps, by submitting ideas that matter to you the most. Here's how it works:
Search existing ideas
Start by searching and reviewing ideas and requests to enhance a product or service. Take a look at ideas others have posted, and add a comment, vote, or subscribe to updates on them if they matter to you. If you can't find what you are looking for,
Post your ideas
Post an idea.
Get feedback from the IBM team and other customers to refine your idea.
Follow the idea through the IBM Ideas process.
Specific links you will want to bookmark for future use
Welcome to the IBM Ideas Portal (https://www.ibm.com/ideas) - Use this site to find out additional information and details about the IBM Ideas process and statuses.
IBM Unified Ideas Portal (https://ideas.ibm.com) - Use this site to view all of your ideas, create new ideas for any IBM product, or search for ideas across all of IBM.
ideasibm@us.ibm.com - Use this email to suggest enhancements to the Ideas process or request help from IBM for submitting your Ideas.
Hi Gino, marking this as not under considering for this year for now, as we will need more information from you on this before it can be considered. Thanks
hi @Gino
This is an interesting idea, and could be applicable only to those k8s workload types that would support scaling down replicas (for example daemonsets and mirror pods are not technically feasible here, and would be handle with node deletion).
We need to build out some requirements here, because there is more to this.
The motivation around managing VPC https://www.ibm.com/docs/en/license-metric-tool?topic=metrics-virtual-processor-core-vpc is good but what kinds of licenses do we want to impact here? CPU Limit or Node VPC? both?
Since Turbonomic considers every workload (pod or vm) at the same level of priority, would this user define a group of workload that they consider "not needed"?
What KPIs would warrant spinning down? Would the user expect to set this as a "threshold" and apply a schedule?
Would the user also automate pods moves so that we could delete a node, which may be the only way to impact the VPC license?
How does the user want to handle spinning this back up? By schedule or expect the user to trigger scaling back out the replicas on their own? What if another node is needed - how would they handle pod pending?
There is a scenario to consider: If the license is allocated to the nodes, then shutting down pods may not help. if you pin pods to nodes (affinity/anti-affinity, node label, taint) and not allow freedom to move the pods that are already running, we may not be able consolidate and suspend nodes if there is even one app pod holding on to that node. Neither could kubernetes suspend that node because the pod running there would not have a compliant node to be scheduled on if the constraints are very restrictive. Turbo can move workloads around but we must respect constraints. Would like your thoughts on how to communicate this because suspending pods would not be enough.