Introducing Argo support for Google Container Engine (GKE), Minikube, and Docker

Mukulika Kapas
Argo Project
Published in
3 min readNov 9, 2017

--

Source: Dilbert cartoon by Scott Adams

Exactly two months ago, we released Argo, an open-source container-native workflow engine for Kubernetes. Each step in an Argo workflow runs as a pod on a Kubernetes cluster. This makes Argo workflows highly scalable, reliable, easy to extend, and portable between Kubernetes clusters. You can read more about it here.

Thanks to everyone, particularly the Kubernetes and CNCF community, for their great feedback and support.

The initial 1.0 release of Argo workflow engine was a packaged solution, which deployed its own Kubernetes cluster from scratch on AWS. We have received several requests to be able to run the Argo workflow engine on an already existing Kubernetes cluster.

We are happy to announce that you can now install and run the new 1.1 release of Argo on existing Kubernetes clusters in AWS, GKE, and MiniKube. We have also created a light weight version that runs as a simple Docker container. So, there are two options to try the 1.1 version of Argo.

  • In less than a minute you can try Argo in a Docker container or Minikube using Argo Lite. Argo Lite provides a flavor of Argo workflow engine on your laptop. You just need to have Docker engine installed.
  • In less than 10 minutes, install Argo on an existing Kubernetes cluster running on GKE, AWS or Minikube and try out the fully functional Argo workflows with artifact management, secret management, and log management features. Get started here!
Argo running on Google Container Engine (GKE)

We also want to let you know that based on your feedback so far, we are hard at work on the following new features.

  • Make Argo a Kubernetes custom resource (CRD) for seamless integration with Kubernetes.
  • Integration with serverless event management frameworks.
  • Certify and support Argo workflow engine on Azure and on-premise private cloud environments.

While continuous integration (CI) remains an important use of workflows, we think that this is just the tip of the iceberg when it comes to Kubernetes. We want to make Argo the new “shell” for getting work done on Kubernetes. Make it as easy to run jobs in the cloud as on your laptop. Imagine a Google-scale supercomputer at your finger tips to do your bidding! Here are some examples where we think container-native workflows could immediately help.

  • Deployment and lifecycle management of distributed containerized apps.
  • Data extraction, migration, and processing (ETL)
  • Image processing
  • Machine learning
  • Modeling and simulations

Argo is an open source project hosted on GitHub and we greatly appreciate your feedback and support. If you like the project, please let us know by starring our GitHub project.

If you are in the San Francisco Bay Area, come and see a live demo of Argo 1.1 at the Bay Area Kubernetes meetup on November 16th.

--

--