Notes on usage of amazon web services for distributed computing
See this link first
See hpc on amazon
Store data in a bucket. You should create a different bucket for each project which allows one to break down storage costs by project.
https://docs.aws.amazon.com/AmazonS3/latest/gsg/CreatingABucket.html
you can also use the aws command line client.
aws s3 mb s3://example-bucket
aws s3 cp help
https://docs.aws.amazon.com/parallelcluster/latest/ug/pre_post_install.html
https://aws.amazon.com/blogs/opensource/aws-parallelcluster/
still an open question