Ask HN: How to cache NFS files across K8s pods on the same node?

I'm using KEDA to schedule jobs on EKS. We store model files on EFS, but if we have 100 jobs they need to read in the same model files 100 times. The nodes themselves have ample disk space and should be able to hold a cache of EFS (files should be considered immutable).

In my case EFS is mounted as a plain NFS drive. Cachefilesd should work but I'd be using a custom AMI at that point. Not terrible. Is there a standard solution here?