During your design cycle and life as a machine learning engineer, you may face occasions where the amount of computation power available at your office is simply not enough, and you can't wait for your IT team to buy you a new server. So, for example, if you can afford 24.48 dollars/hour, you may have a p3.16xlarge with 8 GPUs Nvidia V100, 64 cores, and 488 GB of RAM.
In this section, you will learn about Amazon AWS services that might help you deal with a lack of computing power issues.
You will learn about the following Amazon Cloud Services:
- Elastic Compute Cloud (EC2)
- S3
- SageMaker
This is the service where we create our servers, where you basically create any server to do your work:
Here, you configure stuff like how you want to access your server (normally with a private key):
Here, we configure how much disk space we want:
And here, we configure the ports that are going to be available:
Available ports