Skip to content

Latest commit

 

History

History
32 lines (27 loc) · 2.29 KB

AWS_rough.md

File metadata and controls

32 lines (27 loc) · 2.29 KB

Rough Notes for AWS

Images for EC2

S3

= Mount bucket on EC2 instance: https://medium.com/tensult/aws-how-to-mount-s3-bucket-using-iam-role-on-ec2-linux-instance-ad2afd4513ef

EC2

  • For your own virtual server in the cloud, you would use EC2
  • You base your instance off of an "image" with a particular operating system. e.g. Windows, Debian, etc.
    • All sorts of additional images in the https://aws.amazon.com/marketplace
    • Instances can also have stuff preloaded, we would just need to find nice ones. e.g. probably datascience images/etc.?
    • I think you can add features to the base images easily. e.g. Add CUDA
    • It would be helpful for a research engineer to play around and find good "datascience" etc. images.
  • With it, you only pay for time you are actively running it, and can choose to hibernate/pause.
    • If you pause, I think you only pay for the (dirt-cheap) storage of the image.
  • Also, you could come up with your own container image and store it for quick access with https://aws.amazon.com/ecr/