Integrations with Run:ai¶
The table below summarizes the integration capabilities with various third-party products.
Integration support¶
Support for integrations varies. Where mentioned below, the integration is supported out of the box with Run:ai. With other integrations, our customer success team has previous experience with integrating with the third party software and many times the community portal will contain additional reference documentation provided on an as-is basis.
The Run:ai community portal is password protected and access is provided to customers and partners.
Integrations¶
| Tool | Category | Run:ai support details | Additional Links |
|---|---|---|---|
| Spark | Orchestration | It is possible to schedule Spark workflows with the Run:ai scheduler. For details, please contact Run:ai customer support. | Run:ai customer success community portal: https://runai.my.site.com/community/s/article/How-to-Run-Spark-jobs-with-Run-AI |
| Kubeflow Pipelines | Orchestration | It is possible to schedule kubeflow pipelines with the Run:ai scheduler. For details please contact Run:ai customer support | Run:ai customer success community portal https://runai.my.site.com/community/s/article/How-to-integrate-Run-ai-with-Kubeflow |
| Apache Airflow | Orchestration | It is possible to schedule Airflow workflows with the Run:ai scheduler. For details, please contact Run:ai customer support. | Run:ai customer success community portal: https://runai.my.site.com/community/s/article/How-to-integrate-Run-ai-with-Apache-Airflow |
| Argo workflows | Orchestration | It is possible to schedule Argo workflows with the Run:ai scheduler. For details, please contact Run:ai customer support. | Run:ai customer success community portal https://runai.my.site.com/community/s/article/How-to-integrate-Run-ai-with-Argo-Workflows |
| SeldonX | Orchestration | It is possible to schedule Seldon Core workloads with the Run:ai scheduler. For details, please contact Run:ai customer success. | Run:ai customer success community portal: https://runai.my.site.com/community/s/article/How-to-integrate-Run-ai-with-Seldon-Core |
| Triton | Orchestration | Usage via docker base image | Quickstart inference example |
| Jupyter Notebook | Development | Run:ai provides integrated support with Jupyter Notebooks | Quickstart example: https://docs.run.ai/latest/Researcher/Walkthroughs/quickstart-jupyter/ |
| Jupyter Hub | Development | It is possible to submit Run:ai workloads via JupyterHub. For more information please contact Run:ai customer support | |
| PyCharm | Development | Containers created by Run:ai can be accessed via PyCharm | PyCharm example |
| VScode | Development | - Containers created by Run:ai can be accessed via Visual Studio Code. - You can automatically launch Visual Studio code web from the Run:ai console | VSCode example and VSCode quickstart example. |
| Kubeflow notebooks | Development | It is possible to launch a kubeflow notebook with the Run:ai scheduler. For details please contact Run:ai customer support | Run:ai customer success community portal:https://runai.my.site.com/community/s/article/How-to-integrate-Run-ai-with-Kubeflow |
| Ray | training, inference, data processing. | It is possible to schedule Ray jobs with the Run:ai scheduler. | Run:ai customer success community portal https://runai.my.site.com/community/s/article/How-to-Integrate-Run-ai-with-Ray |
| TensorBoard | Experiment tracking | Run:ai comes with a preset Tensorboard Environment asset. | TensorBoard example. Additional sample |
| Weights & Biases | Experiment tracking | It is possible to schedule W&B workloads with the Run:ai scheduler. For details, please contact Run:ai customer success. | Run:ai Customer success community portal: https://runai.my.site.com/community/s/article/How-to-integrate-with-Weights-and-Biases Additional samples here |
| ClearML | Experiment tracking | It is possible to schedule ClearML workloads with the Run:ai scheduler. For details, please contact Run:ai customer success. | https://runai.my.site.com/community/s/article/How-to-integrate-Run-ai-with-ClearML |
| MLFlow | Model Serving | It is possible to use ML Flow together with the Run:ai scheduler. For details, please contact Run:ai customer support. | Run:ai customer success community portal: https://runai.my.site.com/community/s/article/How-to-integrate-Run-ai-with-MLflow Additional MLFlow sample |
| Hugging Face | Repositories | Run:ai provides an out of the box integration with Hugging Face | |
| Docker Registry | Repositories | Run:ai allows using a docker registry as a Credentials asset | Credentials documentation |
| S3 | Storage | Run:ai communicates with S3 by defining it as a data source asset | Data source documentation |
| Github | Storage | Run:ai communicates with GitHub by defining it as a data source asset | Data source documentation |
| Tensorflow | Training | Run:ai provides out of the box support for submitting TensorFlow workloads | TensorFlow via API or use Workflow submission via user interface |
| Pytorch | Training | Run:ai provides out of the box support for submitting PyTorch workloads | PyTorch via API or use Workflow submission via user interface |
| Kubeflow MPI | Training | Run:ai provides out of the box support for submitting MPI workloads | MPI via API or use Workflow submission via user interface |
| XGBoost | Training | Run:ai provides out of the box support for submitting XGBoost workloads | XGBoost via API or use Workflow submission via user interface |
| Karpenter | Cost Optimization | Run:ai provides out of the box support for Karpenter to save cloud costs | Integration notes with Karpenter can be found here |