Skip to content

Simplified the metrics #3182

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
google-oss-prow merged 13 commits into from
Jul 15, 2025
Merged

Simplified the metrics #3182

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
e5e4c14
Simplified the metrics
kunal-511 Jul 3, 2025
53b21e4
Updated metrics table
kunal-511 Jul 4, 2025
85bc412
Ignoring experimental folder
kunal-511 Jul 8, 2025
60ca6c0
Updated storage fallback
kunal-511 Jul 9, 2025
fce44ce
Readme updated with new meteics numbers
kunal-511 Jul 14, 2025
5f47bec
Updated the component naming
kunal-511 Jul 14, 2025
6b79ae7
Removed old resources table
kunal-511 Jul 14, 2025
0ffb275
Merged resources and kubeflow component version table
kunal-511 Jul 14, 2025
a3df80f
Merge to single table
kunal-511 Jul 15, 2025
95b3670
added kserve web applications resources values
kunal-511 Jul 15, 2025
76e7a3a
Update README.md
juliusvonkohout Jul 15, 2025
842d33b
Update README.md
juliusvonkohout Jul 15, 2025
e32588c
Seperate Dex and OAuth
kunal-511 Jul 15, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .github/workflows/full_kubeflow_integration_test.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,7 +62,7 @@
- name: Install KServe
run: ./tests/kserve_install.sh

#- name: Install Pipelines

Check warning on line 65 in .github/workflows/full_kubeflow_integration_test.yaml

View workflow job for this annotation

GitHub Actions / format_YAML_files 

65:6 [comments] missing starting space in comment
# run: ./tests/pipelines_install.sh

- name: Install Pipelines with SeaweedFS
Expand Down Expand Up @@ -255,7 +255,7 @@
run: |
echo "==== Resource Usage Table ===="
pip3 install -q PyYAML
python3 scripts/generate_resource_table.py
python3 tests/metrics-server_resource_table.py

- name: Collect Logs on Failure
if: failure()
Expand Down
86 changes: 28 additions & 58 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,34 +48,34 @@ All components are deployable with `kustomize`. You can choose to deploy the ent

### Kubeflow Version: Master

This repository periodically synchronizes all official Kubeflow components from the respective upstream repositories. The following matrix shows the git version included for each component:

| Component | Local Manifests Path | Upstream Revision |
| - | - | - |
| Training Operator | applications/training-operator/upstream | [v1.9.2](https://github.com/kubeflow/training-operator/tree/v1.9.2/manifests) |
| Notebook Controller | applications/jupyter/notebook-controller/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/notebook-controller/config) |
| PVC Viewer Controller | applications/pvcviewer-controller/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/pvcviewer-controller/config) |
| Tensorboard Controller | applications/tensorboard/tensorboard-controller/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/tensorboard-controller/config) |
| Central Dashboard | applications/centraldashboard/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/centraldashboard/manifests) |
| Profiles + KFAM | applications/profiles/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/profile-controller/config) |
| PodDefaults Webhook | applications/admission-webhook/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/admission-webhook/manifests) |
| Jupyter Web Application | applications/jupyter/jupyter-web-app/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/crud-web-apps/jupyter/manifests) |
| Tensorboards Web Application | applications/tensorboard/tensorboards-web-app/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/crud-web-apps/tensorboards/manifests) |
| Volumes Web Application | applications/volumes-web-app/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/crud-web-apps/volumes/manifests) |
| Katib | applications/katib/upstream | [v0.18.0](https://github.com/kubeflow/katib/tree/v0.18.0/manifests/v1beta1) |
| KServe | applications/kserve/kserve | [v0.15.0](https://github.com/kserve/kserve/releases/tag/v0.15.0/install/v0.15.0) |
| KServe Models Web Application | applications/kserve/models-web-app | [v0.14.0](https://github.com/kserve/models-web-app/tree/v0.14.0/config) |
| Kubeflow Pipelines | applications/pipeline/upstream | [2.5.0](https://github.com/kubeflow/pipelines/tree/2.5.0/manifests/kustomize) |
| Kubeflow Model Registry | applications/model-registry/upstream | [v0.2.19](https://github.com/kubeflow/model-registry/tree/v0.2.19/manifests/kustomize) |
| Spark Operator | applications/spark/spark-operator | [2.2.0](https://github.com/kubeflow/spark-operator/tree/v2.2.0) |

The following matrix shows the versions of common components used across different Kubeflow projects:

| Component | Local Manifests Path | Upstream Revision |
| - | - | - |
| Istio | common/istio | [1.26.1](https://github.com/istio/istio/releases/tag/1.26.1) |
| Knative | common/knative/knative-serving <br /> common/knative/knative-eventing | [v1.16.2](https://github.com/knative/serving/releases/tag/knative-v1.16.2) <br /> [v1.16.4](https://github.com/knative/eventing/releases/tag/knative-v1.16.4) |
| Cert Manager | common/cert-manager | [1.16.1](https://github.com/cert-manager/cert-manager/releases/tag/v1.16.1) |
This repository periodically synchronizes all official Kubeflow components from the respective upstream repositories. The following matrix shows the git version included for each component along with the resource requirements for each Kubeflow component, calculated as the maximum of actual usage and configured requests for CPU/memory as well as storage requirements from PVCs:

| Component | Local Manifests Path | Upstream Revision | CPU (millicores) | Memory (Mi) | Storage (GB) |
| - | - | - | - | - | - |
| Training Operator | applications/training-operator/upstream | [v1.9.2](https://github.com/kubeflow/training-operator/tree/v1.9.2/manifests) | 3m | 25Mi | 0GB |
| Notebook Controller | applications/jupyter/notebook-controller/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/notebook-controller/config) | 5m | 93Mi | 0GB |
| PVC Viewer Controller | applications/pvcviewer-controller/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/pvcviewer-controller/config) | 15m | 128Mi | 1GB |
| Tensorboard Controller | applications/tensorboard/tensorboard-controller/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/tensorboard-controller/config) | 15m | 128Mi | 0GB |
| Central Dashboard | applications/centraldashboard/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/centraldashboard/manifests) | 2m | 159Mi | 0GB |
| Profiles + KFAM | applications/profiles/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/profile-controller/config) | 7m | 129Mi | 0GB |
| PodDefaults Webhook | applications/admission-webhook/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/admission-webhook/manifests) | 1m | 14Mi | 0GB |
| Jupyter Web Application | applications/jupyter/jupyter-web-app/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/crud-web-apps/jupyter/manifests) | 4m | 231Mi | 0GB |
| Tensorboards Web Application | applications/tensorboard/tensorboards-web-app/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/crud-web-apps/tensorboards/manifests) | | | |

@juliusvonkohout juliusvonkohout Jul 15, 2025

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

where are the values for tensorboards?

| Volumes Web Application | applications/volumes-web-app/upstream | [v1.10.0](https://github.com/kubeflow/kubeflow/tree/v1.10.0/components/crud-web-apps/volumes/manifests) | 4m | 226Mi | 0GB |
| Katib | applications/katib/upstream | [v0.18.0](https://github.com/kubeflow/katib/tree/v0.18.0/manifests/v1beta1) | 13m | 476Mi | 13GB |
| KServe | applications/kserve/kserve | [v0.15.0](https://github.com/kserve/kserve/releases/tag/v0.15.0/install/v0.15.0) | 600m | 1200Mi | 0GB |
| KServe Models Web Application | applications/kserve/models-web-app | [v0.14.0](https://github.com/kserve/models-web-app/tree/v0.14.0/config) | 6m | 259Mi | 0GB |
| Kubeflow Pipelines | applications/pipeline/upstream | [2.5.0](https://github.com/kubeflow/pipelines/tree/2.5.0/manifests/kustomize) | 970m | 3552Mi | 100GB |

@juliusvonkohout juliusvonkohout Jul 15, 2025

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

100 GB for pipelines does not seem correct

| Kubeflow Model Registry | applications/model-registry/upstream | [v0.2.19](https://github.com/kubeflow/model-registry/tree/v0.2.19/manifests/kustomize) | 510m | 2112Mi | 20GB |

@juliusvonkohout juliusvonkohout Jul 15, 2025

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is 20 GB for model registry correct ?

| Spark Operator | applications/spark/spark-operator | [2.2.0](https://github.com/kubeflow/spark-operator/tree/v2.2.0) | 9m | 41Mi | 0GB |
| Istio | common/istio | [1.26.1](https://github.com/istio/istio/releases/tag/1.26.1) | 750m | 2364Mi | 0GB |
| Knative | common/knative/knative-serving <br /> common/knative/knative-eventing | [v1.16.2](https://github.com/knative/serving/releases/tag/knative-v1.16.2) <br /> [v1.16.4](https://github.com/knative/eventing/releases/tag/knative-v1.16.4) | 1450m | 1038Mi | 0GB |
| Cert Manager | common/cert-manager | [1.16.1](https://github.com/cert-manager/cert-manager/releases/tag/v1.16.1) | 3m | 128Mi | 0GB |
| Dex | common/dex | [2.41.1](https://github.com/dexidp/dex/releases/tag/v2.41.1) | 3m | 27Mi | 0GB |
| OAuth2-Proxy | common/oauth2-proxy | [7.7.1](https://github.com/oauth2-proxy/oauth2-proxy/releases/tag/v7.7.1) | 3m | 27Mi | 0GB |
| **Total** | | | **4372m** | **12198Mi** | **134GB** |



## Installation

Expand Down Expand Up @@ -707,36 +707,6 @@ The hooks will run automatically on `git commit`. You can also run them manually
pre-commit run
```

## Resource Usage by components

The following table shows the resource requirements for each Kubeflow component, calculated as the maximum of actual usage and configured requests for CPU/memory, plus storage requirements from PVCs:

| Component | CPU (cores) | Memory (Mi) | Storage (GB) |
|-----------|-------------|-------------|--------------|
| Dex + OAuth2-Proxy | 3m | 27Mi | 0GB |
| Cert Manager | 3m | 130Mi | 0GB |
| Istio | 850m | 2464Mi | 0GB |
| Katib | 4m | 107Mi | 3GB |
| Kubeflow Core | 17m | 828Mi | 0GB |
| KServe | 600m | 1200Mi | 0GB |
| Metadata | 10m | 225Mi | 40GB |
| Model Registry | 500m | 2048Mi | 0GB |
| Pipelines | 770m | 3276Mi | 60GB |
| Spark | 5m | 36Mi | 0GB |
| Training | 2m | 26Mi | 0GB |
| Other | 1615m | 1698Mi | 36GB |
| **Total** | **4379m** | **12065Mi** | **139GB** |

### Resource Notes

- **CPU values** represent the maximum of actual observed usage and configured resource requests
- **Memory values** represent the maximum of actual observed usage and configured resource requests
- **Storage values** represent the total PVC allocations from manifest files
- Components with high resource requests (like Istio, KServe) may be over-provisioned compared to actual usage
- Storage requirements are persistent and represent the minimum disk space needed

Use this as a reference when planning your Kubeflow installation to allocate appropriate resources and decide which components to enable based on your available infrastructure.

## Frequently Asked Questions

- **Q:** What versions of Istio, Knative, Cert-Manager, Argo, ... are compatible with Kubeflow?
Expand Down
Loading
Loading