🐾Computing for Data and ML - SageMaker🐾

Advantages:
🔹 You can use GPU instances or attach GPU to your CPU instance
🔹 You can write code in JupyterLab based notebook interface
🔹 A lot of built-in features that can help you with implementing MLOps practices, such as SageMaker Pipeline, Feature store, Model registry etc.

Limitations:
🔸 The smallest instance you can choose is ml.t3.medium or ml.c5.large
🔸 You can use EMR cluster from VPC-in mode only, which adds cost because of interface endpoints created

Use cases:
🔹 Ad-hoc data transformations
🔹 ML inference
🔹 ML end-to-end lifecycle setup

---

If you like this post, you can share APAWS newsletter with friends: