125 points by ml_engineer 6 months ago flag hide 10 comments
k8s-master 6 months ago next
I've been experimenting with Kubernetes for months and this guide has helped me incorporate ML models. Super helpful!
deployment-nerd 6 months ago next
@k8s-master glad to hear that! Kubernetes is definitely the way to go for production-ready ML models.
mlops-guru 6 months ago prev next
This is an excellent guide! I've been looking for a comprehensive resource on deploying ML models using Kubernetes. Bravo!
cloud-tech-pro 6 months ago next
@mlops-guru you're welcome! It took a lot of research to compile this information. I'm glad it could help you!
oracle-of-devops 6 months ago prev next
Any recommendations for monitoring ML models deployed using Kubernetes? I've been looking for a reliable solution.
kuber-cadet 6 months ago next
@oracle-of-devops Check out Prometheus with Grafana. It's an open-source solution designed for monitoring containers, and it works really well with Kubernetes.
data-whisperer 6 months ago prev next
Any gotcha's when deploying ML models with Kubernetes? I'd like to avoid some common pitfalls.
continuity-drone 6 months ago next
@data-whisperer Keep an eye out for resource management. Monitor you container resources closely, especially GPU resources if you use them!
nimble-nimrod 6 months ago prev next
This is a solid article, but where are the recommendations for continuous integration and delivery pipelines?
wise-wargamer 6 months ago next
@nimble-nimrod Jenkins X works wonderfully for continuous delivery on Kubernetes. Give it a try!