kepler

kepler copied to clipboard

Create ephemeral cluster with k3d/k3s

2

##What Added an option for k3d/k3s clusters ##Why An alternative flavor to kind ##How Based on the existing cluster/cluster-up/kind/common.sh script logic, with the necessary minor changes to facilitate k3d ##Tested...

akyriako

Ephemeral Pod is not always found by Kepler

7

**Describe the bug** A clear and concise description of what the bug is. **To Reproduce** I used ARC to create a self-hosted runner for GitHub Actions. It runs for 10-15s...

nikimanoledaki

Make `pod_total_*` metrics persistent/survive a Kepler restart

8

**A clear and concise description of what the problem is.** The `pod_total_*` metrics represent the total accumulated over all sample since Kepler started monitoring the particular pod. When restarting the...

williamcaban

Kepler is very slow to start

6

**Describe the bug** Before the PR #225, Kepler was taking 1.18ms to start. But now, it is taking 40.17s to start Although It might be related to the model server,...

marceloamaral

Update Grafana dashboards with the new container metrics

13

**Why this PR is needed?** PR #287 will update Prometheus metrics and affect the current Grafana dashboard. Where the new metrics will report energy per container and have more meaningful...

marceloamaral

Improve shared memory GPU power estimate using perf counters

3

**Is your feature request related to a problem? Please describe.** Current GPU power estimate is based on memory footprint. This model doesn't account for factors such as GPU instructions, temperature,...

rootfs

Enhance CI with basic integration test

2

**Is your feature request related to a problem? Please describe.** So far, we just verify by `until kubectl get svc --all-namespaces ...` to make it better, we'd better have a...

SamYuan1990

aggregate energy consumption from Pod level to higher level

9

**Is your feature request related to a problem? Please describe.** Pod level energy reporting is useful on a per node basis. At the cluster level, high level aggregation is more...

rootfs

5s is not enough for e2e test on Unit test Mac

5

**Describe the bug** Regarding this PR https://github.com/sustainable-computing-io/kepler/pull/225, the Unit test Mac keeps failing on e2e test (while the other unit tests were fine) when fixing first waiting time to 5s....

sunya-ch

Detect kepler is running inside VM

4

followup, we will use kustomization to build different manifest .. Signed-off-by: jichenjc

jichenjc