search

GSTT-CSC / MLOps

Framework for building ML apps
GNU General Public License v3.0
9 stars 5 forks source link

Move to nvidia deepops/k8s backend #5

Open laurencejackson opened 3 years ago

laurencejackson commented 3 years ago

Moving to use the nvidia DeepOps k8s backend will enable more control over resource management and enable us to scale effectively. This should integrate well with the experimental k8s backend for mlflow projects.

laurencejackson commented 2 years ago

Seldon core looks like a nice alternative way to deploy to K8s if the mlflow implementation is still in beta. Looks like opportunities to integrate both too. Simple microk8s instance might be better than full-blown deepops for now.

https://docs.seldon.io/projects/seldon-core/en/v1.1.0/workflow/quickstart.html

https://docs.seldon.io/projects/seldon-core/en/v1.1.0/servers/mlflow.html