Menu
Apache MXNet on AWS
Developer Guide

Step 1: Launch an EC2 Instance

Launch an EC2 instance using the Ubuntu version of the Deep Learning AMI. The AMI includes everything that you need to set up an instance and test sample MXNet code using Python.

Note

Although the Deep Learning AMI is available for both Ubuntu and Amazon Linux, in this exercise, we use the Ubuntu version.

Launch the instance

  1. In the Amazon EC2 console, launch an instance. For step-by-step instructions, see Launching an AWS Marketplace Instance in the Amazon EC2 User Guide for Linux Instances. As you follow the steps, use the following values:

    • On the Choose an Amazon Machine Image (AMI) page, choose the AWS Marketplace tab, and search for the Deep Learning AMI Ubuntu Version AMI.

    • On the Choose an Instance Type page, choose the c4.4xlarge instance type.

    • On the Configure Instance Details page, do the following:

      • From the Network drop-down, choose your VPC.

      • From the Auto-assign Public IP list, choose Enable

    • On the Add Storage page, choose the default storage size. The AMI uses about 38 GB of disk space.

    • On the Add Tags page, add one tag with the key Name and any value.

    • On the Configure Security Group page, choose Add Rule, and then add the following custom Transmission Control Protocol (TCP) rule.

      Type : Custom TCP Rule

      Protocol: TCP

      Port Range: 8888

      Source: Anywhere (0.0.0.0/0,::/0)

      Note

      In this exercise, you also set up a Jupyter notebook server on the EC2 instance. After you set up the environment, the server for the Jupyter notebook starts on port 8888.

  2. Wait for the instance to be ready. You can verify the status of the instance in the Amazon EC2 console.

Next Step

Step 2: Connect to the EC2 Instance