You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am working on a project to create a Kubernetes scheduler that makes decisions based on the power efficiency of each node. To achieve this, I need accurate energy consumption metrics for each node in the cluster.
While using Kepler, I can aggregate energy consumption for individual pods running on a node. However, I have noticed that there are two additional namespaces—system and kernel—that account for the energy consumption of system-level processes and the kernel across the entire node.
The problem is that I cannot find a way to disaggregate the energy consumption of these namespaces into per-node metrics. This makes it difficult to obtain a total energy usage figure for each node that includes both pod-level and system/kernel-level energy consumption.
My question is:
Is there a way to measure the total energy consumption per Kubernetes node, including all processes (pods, system, kernel, etc.)?
If not, are there any workarounds or planned features in Kepler that might help in capturing this data more accurately at the node level?
Thank you for your help!
The text was updated successfully, but these errors were encountered:
Issue Description
I am working on a project to create a Kubernetes scheduler that makes decisions based on the power efficiency of each node. To achieve this, I need accurate energy consumption metrics for each node in the cluster.
While using Kepler, I can aggregate energy consumption for individual pods running on a node. However, I have noticed that there are two additional namespaces—system and kernel—that account for the energy consumption of system-level processes and the kernel across the entire node.
The problem is that I cannot find a way to disaggregate the energy consumption of these namespaces into per-node metrics. This makes it difficult to obtain a total energy usage figure for each node that includes both pod-level and system/kernel-level energy consumption.
My question is:
Thank you for your help!
The text was updated successfully, but these errors were encountered: