Skip to content

Latest commit

 

History

History

model-server

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
src
src
 
 
README.md
README.md
 
 
massdriver.yaml
massdriver.yaml
 
 

README.md

k8s-model-server

An example machine-learning model server running on GKE with GPUs.

TODO

  • Integrate with GCS and GCS Fuse

An example application using this template can be found here.

Massdriver kubernetes-deployment Template

The kubernetes-deployment template will run your application on AWS, GCP, or Azure Kubernetes.

Files:

  • a terraform module is included that configurings IAM permissions and sets up you environment variables. You likely do not need to modify these files. This module is simply rigging code to integrate with Massdriver Cloud.
  • a helm chart has been created to run a kubernetes deployment. This Helm chart is a great getting started point for deploying to Kubernetes. Feel free to modify the chart to customize your application deployment.
  • the massdriver.yaml controls the UI to expose for configuring your application and its dependencies. By default there are a lot of fields in your params section, feel free to remove fields that you do not want exposed in your configuration form in Massdriver Cloud. Values that you do not want to change (e.g.: your image repository) can be hard coded in the values.yaml file.