AWS offers different types of Notebook instances: SageMaker Studio, SageMaker Notebook Instances, and SageMaker Studio Lab. SageMaker Notebook Instances offer the most flexibility in terms of installing local software via conda/mamba or launching from custom containers. [Amazon SageMaker Notebook Instance](https://docs.aws.amazon.com/sagemaker/latest/dg/nbi.html) allows for 'on the fly' machine resizing and notebook scheduling, as well as not worrying about resource availability, but they run in a managed environment and offer less flexibility for installing custom software. Most machine-learning related software are pre-installed, but these can be hard to use for a lot of bioinformatic tasks where you need to install CLI tools with conda.
0 commit comments