Requirements

To use the MATLAB Deep Learning Container, you need:

Additionally, if you have a Concurrent license type, you must supply the port number and DNS address of the network license manager when you run the container. Add an option of the following form to the docker run command when you start the container.

-e MLM_LICENSE_FILE=27000@MyLicenseServer

Costs

You are responsible for the cost of the Amazon Web Services used when you create a cluster using this guide. Resource settings, such as instance type, affect the cost of deployment. For cost estimates, see the pricing pages for each AWS service you are using. Prices are subject to change.

Prepare Your AWS Account

If you do not have an Amazon Web Services account, create one at https://aws.amazon.com by following the on-screen instructions. Create a key pair using the Amazon EC2 Console. For more information, see Amazon EC2 Key Pairs.

Launch Docker Host Instance

Log in to your Amazon Web Services Console. Click Services > Compute > EC2 and launch an instance.

On the Launch instance page, name your instance and search for the NVIDIA GPU-Optimized AMI under Application and OS Images, select it and confirm. This Amazon Machine Image (AMI) is designed for use with NVIDIA GPU Cloud to take advantage of the Volta GPUs available in P3 instances.

Configure your instance type, network settings, and storage as needed. Note that not all Availability Zones offer P3 instances. Your Availability Zone is defined during setup of your virtual private cloud (VPC).

If necessary, choose or create appropriate Security Groups for your instance.

Select the appropriate key pair option and launch your instance. Make sure that you have access to your private key so you can log in to your instance.

After your instance launches successfully, open the Instances page from the navigation bar on the left. Locate the instance corresponding to your instance name. When the Status check field shows that all the checks pass, you can connect to your instance.

Connect to Instance

Click the Instance ID of your instance to go to the Instance summary page, which shows the public IPv4 DNS address of your instance. Connect to your instance by using SSH from your client machine with your private key, using PuTTY or another SSH client. If you use PuTTY, add the private key under Connection > SSH > Auth. If you use a command-line interface, add the private key using the tag -i "path/to/key.pem". The default username to connect to your EC2 instance is ubuntu. The address to connect to the instance has this form:

ubuntu@ec2-public-ipv4-address.amazonaws.com

Pull Container

After connecting to the instance, pull the container to download the container image onto the Docker host instance, the machine on which the container is to run. You need to do this only once per EC2 instance.

You can copy the pull command for the container image release from the NVIDIA GPU Cloud Container Catalog. In the Tags section, locate the container image release that you want to run. In the Pull column, click the icon to copy the docker pull command. The command has the form:

docker pull nvcr.io/partners/matlab:r20XYz

Paste the docker pull command into your SSH client, and run the command on your EC2 instance. You do not need to log in to the NVIDIA Container Catalog to pull the container image.

Running the docker pull command downloads the MATLAB container image onto the host EC2 machine. Downloading and extracting the large container image can take some time.

Run Container

Run the MATLAB Deep Learning Container using a command in this form:

docker run -it --rm -p 5901:5901 -p 6080:6080 --gpus all --shm-size=512M nvcr.io/partners/matlab:r20XYz

Ensure the last part of the run command matches the MATLAB release you want to use.

The option -p hostport:containerport maps ports from the Docker host instance to ports inside the container so that you can connect to the container desktop. The container uses the ports 5901 (for VNC connection) and 6080 (for web browser connection). If you are deploying multiple containers on the same host instance, you must increment the host ports until you find a free port. For example:

-p 5902:5901 -p 6081:6080

The MATLAB Deep Learning Container is now running on your EC2 machine.

Run MATLAB in Container

Access MATLAB in the container in one of three ways:

http://localhost:6080

matlab

localhost:1 

matlab

matlab

Test Container Using Deep Learning Example

MATLAB supports training a single network in parallel using multiple GPUs. To enable multi-GPU training in the MATLAB Deep Learning Container, use the trainingOptions function to set 'ExecutionEnvironment' to 'multi-gpu'.

Train your network using the trainNetwork function. MATLAB opens a parallel pool of workers on all available GPUs. To select only specific GPUs for training, use gpuDevice. For further information, see Select Particular GPUs to Use for Training (Deep Learning Toolbox).

To test your container, you can run the Create Simple Deep Learning Neural Network for Classification (Deep Learning Toolbox) example. To try this example, double-click the file MNISTExample.mlx in the Current Folder pane in the MATLAB startup folder. To run this example on all available GPUs, in the trainingOptions function, set the 'ExecutionEnvironment' to 'multi-gpu'.

Close Container Session

To close the container session, type exit from the container terminal. Doing so stops the container and removes it. No processes or data are saved by default when the container is closed unless you have saved data in the cloud by mounting cloud storage, as described in Share Data with Containers.

Terminate Instance

When you no longer need your instance, terminate it. Terminating the instance deletes all the data associated with it. To terminate your EC2 instance, go to the Instance summary page and click Terminate instance in the Instance drop-down list.

Additional Information