kepler

kepler copied to clipboard

Bump up to golang 1.21

5

Bump up to golang 1.21 _Originally posted by @SamYuan1990 in https://github.com/sustainable-computing-io/kepler/pull/924#discussion_r1327891431_

SamYuan1990

Add context switch count to Kepler ebpf metrics

2

### What would you like to be added? Since kepler is already in the context switch tracepoint, we can add the number of context switches of a process to the...

rootfs

Bumps [github.com/NVIDIA/go-nvml](https://github.com/NVIDIA/go-nvml) from 0.12.0-1 to 0.12.0-3. Commits 7d03741 Merge pull request #104 from ArangoGutierrez/ghactions 6d11ae6 Merge pull request #105 from klueska/init-refcount a0e9de3 Add a refcount to dl load() and close()...

dependabot[bot]

CI enhancement, add logic when we need to test build logic

5

Currently our CI logic is just based on pr or push. hence when we making changes on dockerfile or CI as base image github action, the test is not triggered,...

SamYuan1990

Enable process metrics in CI

7

### What happened? https://github.com/sustainable-computing-io/kepler/issues/1153 caused some issue , we should enable this in the CI in order to make sure it's not break ### What did you expect to happen?...

jichenjc

Kernel consumption too high

25

### What happened? The measured consumption for the kernel is much too high, beyond what the machine is physically capable of.  ### What did you expect to happen? Correct...

jochen-schuettler

Documentation: Integration of Telegraf in Kepler

2

### What would you like to be added? Document covering about the instructions of integrating Telegraf in Kepler where both of them can point to the same instance of the...

Shobhij

Integrate Kepler with Graphite ?

5

Hello I have been using Kepler for some time and it has worked fine, however, due to some external reasons, I had to move from Prometheus + Grafana stack into...

annis-souames

OpenTelemetry deployment or API integration

21

For deployment integration, evaluate the architecture of metrics -> telemetry adapter For API integration, evaluate telemetry client scalability in kepler

rootfs

In `GetGPUPower()`, Kepler should use the per GPU Power consumption to calculate the process's how much power consumed by the process

1

### What would you like to be added? Here is the [current implementation](https://github.com/sustainable-computing-io/kepler/blob/main/pkg/model/estimator/local/ratio.go#L168). It takes into account of all the GPU power from every devices and attributes regardless of their...

rootfs