opendatahub-io / ai-edge

ODH integration with AI at the Edge usecases
Apache License 2.0
9 stars 19 forks source link

Add on-prem S3 to the use case as the default for s3 storage for any edge demo #70

Open LaVLaS opened 1 year ago

LaVLaS commented 1 year ago

Instead of requiring paid S3 access for anyone who wants to run this demo, we should integrate an on-prem S3 service, like minio, to provide a self-contained demo environment that does not rely on external services. This would also help reduce any potential credential leaks when mutiple people attempt to deploy this use case in their environment.

As a parallel task, any AWS s3 buckets that are required for prepping an environment for the use should only be read-only access and treated a source repository for seeding the initial artifacts for a demo

grdryn commented 1 year ago

This might make #57 unnecessary.

LaVLaS commented 1 year ago

This might make #57 unnecessary.

I think we can host read-only, anonymous buckets for artifacts to be consumed as the initial seed for any use case. I think this would consist of binary objects that are not easily versioned by git. This issue is an attempt to ensure that a running use case does not require external S3 for these small use cases

adelton commented 1 year ago

Is this for the model storage, or for the ACM Observability?