Project Meteor
Available at: https://shower.meteor.zone
Similar to how a meteoroid enters the atmosphere, becomes visible as a meteor and finally hits the Earth’s surface as a meteorite, with this initiative we aim to create better visibility and impact of the tools, processes and expertise developed by the AICoE through project Meteor.
This project is being driven by members of the Artificial Intelligence Center of Excellence (AICoE), part of Red Hat’s Office of the CTO. This team is focused on applying AI to Red Hat’s core business and services through open source development that enables user’s needs and fosters open collaboration in the data science and machine learning space.
Project Meteor is a combined effort across the AICoE to provide a single tool for data scientists and other users where they can interact with, explore and leverage all of our services, tools and technologies for developing intelligent applications.
Installation
Prerequisites
- OpenShift Pipelines
- OpenShift 4.7+
Steps
Meteor can be installed as an Operator to any OpenShift cluster.
Since we’re still in early stages of development, we require users to deploy a custom Operator Catalog to allow faster iterations. In future the Meteor operator will be included as a Community Operator to standard OpenShift operator catalog.
-
Deploy CatalogSource
oc apply -f https://raw.githubusercontent.com/AICoE/meteor-operator/main/config/olm/catalogsource.yaml
-
Subscribe to the Operator. This can be done either via UI - OpenShift web console under Administrator perspective -> Operators -> Operator Hub -> search for “Meteor” and click Install or manually via CLI:
oc new-project aicoe-meteor oc apply -f https://raw.githubusercontent.com/AICoE/meteor-operator/main/config/olm/operatorgroup.yaml oc apply -f https://raw.githubusercontent.com/AICoE/meteor-operator/main/config/olm/subscription.yaml
-
Deploy Meteor Shower
kustomize build github.com/AICoE/meteor/manifests | oc apply -f -
-
Deploy pipelines provided by Project Meteor and Meteor auxillary deployments
kustomize build github.com/AICoE/meteor-operator/config/pipelines | oc apply -f - kustomize build github.com/AICoE/meteor-operator/config/pushgateway | oc apply -f -
All steps are subject to change in future updates. We aim to bake in steps 3 and 4 into the operator itself to provide a single click deployment scheme.
Community
To keep up with this project’s progress, check out Meteor on the Operate First YouTube channel.