I am a fan of Gearman to schedule and dispatch distributed jobs, Redis as a collaborative blackboard, and GlusterFS to share models across multiple systems and make bulk data available across the entire system (usually referenced in the blackboard as a pathname). Source: about 1 year ago

If you're not relying on support, then I would probably standardize on the latest packages available from gluster.org. Source: almost 3 years ago

Cloud Storage: blog storage for static assets and media files. - Source: dev.to / 6 months ago

Preevy includes built-in support for saving profiles on AWS S3 and Google Cloud Storage. You can also store the profile on the local filesystem and copy it manually before running Preevy - we won't show this method here. - Source: dev.to / 6 months ago

Google Cloud Storage{:target="_blank"} is a globally distributed object storage service offered by Google Cloud Platform. They provide trustworthy and scalable databases for storing large amounts of blob data. They also provide a way to optimize cost and performance with different storage classes and pricing options. - Source: dev.to / 11 months ago

Google Cloud Storage - https://cloud.google.com/storage/. - Source: dev.to / 12 months ago

Also, in terms of packing a pre-trained model you will probably want to puts weights, biases etc into S3 or similar object storage (https://cloud.google.com/storage etc) and load it on application start. Source: about 1 year ago