AWS Reference Implementation #24
Conversation
reference_implementations/aws/online/aws_sagemaker/ingest_data_to_fs.py
Outdated
Show resolved
Hide resolved
| project = "ai-deployment-bootcamp" | ||
| region = "us-east-1" | ||
| prefix = "poc" | ||
| infra_env = "dev" |
There was a problem hiding this comment.
Does this work similar to the env var on the GCP implementation (see also here)? i.e. if I change the env to something like env-marcelo I can deploy a whole parallel architecture without affecting whatever has been deployed before?
There was a problem hiding this comment.
Yes, they're similar, with the difference that in AWS we don't have separate projects and we can use infra_env and append it to resource names or use it in other properties (e.g. Environment tags) to ensure isolation from other resources.
There was a problem hiding this comment.
Is this already being done, or do I need to manually append it in case I need to do that?
There was a problem hiding this comment.
No, I haven't done it so you need to append your env name to resources
There was a problem hiding this comment.
Can you do it here as well? It was an important part of the GCP implementation as it allows the participants to spin up multiple test pipelines at the same time. Doing it beforehand for them makes it very handy.
This PR adds reference implementation for online and offline inference.