AWS is undoubtedly one of the most popular choices for machine learning cloud infrastructure. It offers wide range of hardware configuration (CPU and GPU) and extensive datacenter footprint across the globe. AWS also offers its spare capacity as pre-emptible spot instances, at a fraction of their retail cost. These spot instances take the pinch out of otherwise very expensive GPU cloud machines, and present themselves as an excellent choice for Machine Learning developers and Data Scientists.
Clouderizer has in-built deep integration with AWS. You can link your Clouderizer account with your AWS account, which allows you to create, configure, manage and delete AWS resources (including spot instances) from Clouderizer console itself. Below are the steps to do so
- Login to your AWS account (Sign Up in case you don’t have one already)
- Click on the name of your account (it is located in the top right corner of the console). In the expanded drop-down list, select My Security Credentials.
- Click on Get Started with IAM Users button.
- On next screen, click Add user button to add a new user for Clouderizer.
- Give a username e.g. clouderizer_user and set Access type to Programmatic access. Press Next.
- Select option Attach existing policies directly. Then from list below search and select AdministratorAccess policy. Press Next.
- Confirm the settings and press Next. Now you should see the success message with Access key ID and Secret access key. Note down both the values (you will need to press Show to view Secret access key).
- Now login to your Clouderizer console and go to Settings->Cloud Settings
- Enter the Access Key ID and Secret Access Key from step above, and press Update.
Thats it. Your Clouderizer account is now integrated with your AWS account. You can now create projects with AWS SPOT instances, get best recommended bidding price for your configuration, across regions, start and stop those machines from Clouderizer console itself.
In case you have any questions related to AWS integration with Clouderizer, feel free to discuss it on Clouderizer Community. We will try our best to get back to you as soon as possible.
Originally published at blog.clouderizer.com.
Source: Deep Learning on Medium