kepler
kepler copied to clipboard
Kepler (Kubernetes-based Efficient Power Level Exporter) uses eBPF to probe performance counters and other system stats, use ML models to estimate workload energy consumption based on these stats, and...
so far if there no gpu device, the code still go through with gpu logic for example: `podGPUDelta` as we have `*enableGPU`, if `GPU` is disabled, can we skip gpu...
we have an issue as we installed bcc at container, but if host vm/instance don't bcc lib installed, and bcc been used as volume mount of pod, the pod will...
**Is your feature request related to a problem? Please describe.** Similar to #252, energy spent in network irq handling should be accounted and attributed. **Describe the solution you'd like** There...
**Is your feature request related to a problem? Please describe.** Set up role based auto merge. Currently we have [assigned roles](https://github.com/sustainable-computing-io/kepler/blob/main/Contributors.md). The CI can take this info and auto merge...
**Is your feature request related to a problem? Please describe.** Currently the attribution to Pods is based the Pods CPU/memory activities. The processes in the Pods could have disk I/O...
**Is your feature request related to a problem? Please describe.** - remove post integration test - image build from base to kepler **Describe the solution you'd like** n/A **Describe alternatives...
to fix #252
**Describe the bug** All Pods return the same value in a kind cluster. **To Reproduce** On a MacOS, I created a Vagrant VM with VirtualBox as a driver to run...
**Describe the bug** Kepler panics when it does not find kernel-headers (`linux-headers-xyz`) with the same version as the kernel version (`uname -r`). **To Reproduce** kernel version: ``` uname -r 5.10.77...
**Is your feature request related to a problem? Please describe.** Try to add commit message check for each pr. **Describe the solution you'd like** https://www.kubernetes.dev/docs/guide/pull-requests/#commit-message-guidelines - Try to keep the...