Skip to content

Getting started#

Can my data be recovered once I've terminated my instance?#

Warning

We cannot recover your data once you've terminated your instance! Before terminating an instance, make sure to back up all data that you want to keep.

If you want to save data even after you terminate your instance, create a persistent storage file system .

Note

The persistent storage file system must be attached to your instance before you start your instance. The file system cannot be attached to your instance after you start your instance.

When you create a file system, a directory with the name of your file system is created in your home directory. For example, if the name of your file system is PERSISTENT-FILE-SYSTEM, the directory is created at /home/ubuntu/PERSISTENT-FILE-SYSTEM. Data not stored in this directory is erased once you terminate your instance and cannot be recovered.

Can I pause my instance instead of terminating it?#

It currently isn't possible to pause (suspend) your instance rather than terminating it. But, this feature is in the works.

Until this feature is implemented, you can use persistent storage file systems to imitate some of the benefits of being able to pause your instance.

Do you support Kubernetes (K8s)?#

Yes. You can install and use Kubernetes, also known as K8s, on on-demand instances and 1-Click Clusters .

Additionally, Managed Kubernetes and Pre-Installed Kubernetes are available for 1-Click Clusters and Reserved Cloud .

Managed Kubernetes includes:

  • Kubernetes installation and upgrades.
  • Control plane maintenance and high-availability.
  • NVIDIA GPU Operator installed and configured to deploy and manage NVIDIA GPUs in a Kubernetes cluster.
  • Detecting node failures, node pool adjustment and failed hardware replacement.
  • Gathering chassis and cluster metrics and proactive monitoring.

See our Managed Kubernetes Product Outline to learn more.

Why can't my program find the NVIDIA cuDNN library?#

Unfortunately, the NVIDIA cuDNN license limits how cuDNN can be used on our instances.

On our instances, cuDNN can only be used by the PyTorch® framework and TensorFlow library installed as part of Lambda Stack.

Other software, including PyTorch and TensorFlow installed outside of Lambda Stack, won't be able to find and use the cuDNN library installed on our instances.

Tip

Software outside of Lambda Stack usually looks for the cuDNN library files in /usr/lib/x86_64-linux-gnu. However, on our instances, the cuDNN library files are in /usr/lib/python3/dist-packages/tensorflow.

Creating symbolic links, or "symlinks," for the cuDNN library files might allow your program to find the cuDNN library on our instances.

Run the following command to create symlinks for the cuDNN library files:

for cudnn_so in /usr/lib/python3/dist-packages/tensorflow/libcudnn*; do
  sudo ln -s "$cudnn_so" /usr/lib/x86_64-linux-gnu/
done

How do I open JupyterLab on my instance?#

To open JupyterLab on your instance:

  1. In the GPU instances dashboard , find the row for your instance.
  2. Click Launch in the Cloud IDE column.

Tip

Watch Lambda's GPU Cloud Tutorial with JupyterLab video on YouTube to learn more about using JupyterLab on Lambda GPU Cloud instances.

How do I upgrade Python?#

Warning

Upgrading Python, that is, replacing the preinstalled Python version with a newer version, will break your instance.

Instead of upgrading Python, you should install your desired version of Python alongside the preinstalled version, and use your desired version in a virtual environment.

To install another version of Python alongside the preinstalled version:

  1. Run the following command:

    sudo apt -y update && sudo apt -y install pythonVERSION-full

    Replace VERSION with the Python version you want to install, for example, 3.13. Make sure -full is appended to the Python version, otherwise, you won't have the venv module needed to create Python virtual environments.

    As a complete example, if you want to install Python version 3.13, run:

    sudo apt -y update && sudo apt -y install python3.13-full

  2. Run the following command to create a Python virtual environment:

    pythonVERSION -m venv VENV-NAME

    Replace VERSION with the Python version you installed in the previous step. Replace VENV-NAME with the name you want to give your virtual environment.

  3. Run the following command to activate the virtual environment:

    source VENV-NAME/bin/activate

    Replace VENV-NAME with the name you gave your virtual environment.

    As a complete example, if you want to create a virtual environment named my-virtual-environment using Python version 3.13 (installed in the example in the previous step), run:

    python3.13 -m venv my-virtual-environment
source my-virtual-environment/bin/activate

  4. Run the following command to confirm that your virtual environment is using your desired Python version:

    python --version

Can I upgrade to the latest Ubuntu release?#

Warning

Do not run sudo do-release-upgrade or try to upgrade to the latest Ubuntu release. Doing so will break JupyterLab and unless you have SSH access to your instance, Lambda Support won't be able to help you recover your data.

JupyterLab on our instances is configured and tested for the preinstalled version of Python. Upgrading to the latest Ubuntu release will replace the preinstalled version of Python and make JupyterLab inaccessible.

Is it possible to use more than one SSH key?#

It's possible to allow more than one SSH key to access your instance. To do so, you need to add public keys to ~/.ssh/authorized_keys. You can do this with the echo command.

Tip

You can also import SSH keys from GitHub.

Note

This FAQ assumes that you've already generated another SSH key pair, that is, a private key and a public key.

Public keys look like this:

ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIK5HIO+OQSyFjz0clkvg+48YAihYMo5J7AGKiq+9Alg8 user@hostname

SSH into your instance as you normally do and run:

echo 'PUBLIC-KEY' >> ~/.ssh/authorized_keys

Replace PUBLIC-KEY with the public key you want to add to your instance. Make sure to keep the single quotes (' ').

You should now be able to log into your instance using the SSH key you just added.

Tip

You can make sure the public key has been added by running:

cat ~/.ssh/authorized_keys

The last line of output should be the public key you just added.

What SSH key formats are supported?#

You can add SSH keys in the following formats using the dashboard or the Cloud API:

  • OpenSSH (the format ssh-keygen uses by default when generating keys)
  • RFC4716 (the format PuTTYgen uses when you save a public key)
  • PKCS8
  • PEM

Note

  • OpenSSH keys look like:

    ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIK5HIO+OQSyFjz0clkvg+48YAihYMo5J7AGKiq+9Alg8 foo@bar

  • RFC4716 keys begin with:

    ---- BEGIN SSH2 PUBLIC KEY ----

  • PKCS8 keys begin with:

    -----BEGIN PUBLIC KEY-----

  • PEM keys begin with, for example:

    -----BEGIN RSA PUBLIC KEY-----

How long does it take for instances to launch?#

Single-GPU instances usually take 3-5 minutes to launch.

Multi-GPU instances usually take 10-15 minutes to launch.

Note

JupyterLab can take a few minutes after an instance launches to become accessible.

Note

Billing starts after the instance successfully boots and passes health checks.

What network bandwidth does Lambda GPU Cloud provide?#

Note

Some sites limit transfer speeds. This is known as bandwidth throttling.

Lambda GPU Cloud doesn't limit your transfer speeds but can't control other sites' use of bandwidth throttling.

Further, real-world network bandwidth depends on a variety of factors, including the total number of connections opened by your applications and overall network utilization.

Utah, USA region (us-west-3)#

The bandwidth between instances in our Utah, USA region (us-west-3) can be up to 200 Gbps.

The total bandwidth from this region to the Internet can be up to 20 Gbps.

Texas, USA region (us-south-1)#

The bandwidth between instances in our Texas, USA region (us-south-1) can be up to 200 Gbps.

The total bandwidth from this region to the Internet can be up to 20 Gbps.

Note

We're in the process of testing the network bandwidth in our other regions.

How do I learn my instance's private IP address and other info?#

You can learn your instance's private IP address with the ip command.

You can learn what ports are open on your instance with the nmap command.

Learn your instance's private IP address#

To learn your instance's private IP address, SSH into your instance and run:

ip -4 -br addr show enp5s0

The above command will output, for example:

enp5s0           UP             10.19.51.71/20

In the above example, the instance's private IP address is 10.19.51.71.

Tip

If you want your instance's private IP address and only that address, run the following command instead:

ip -4 -br addr show enp5s0 | grep -Eo '(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)'

The above command will output, for example:

10.19.51.71

Learn what ports on your instance are publicly accessible#

You can use Nmap to learn what ports on your instance are publicly accessible, that is, reachable over the Internet.

Note

The instructions, below, assume you're running Ubuntu on your computer.

First, install Nmap on your computer (not on your instance) by running:

sudo apt install -y nmap

Next, run:

nmap -Pn INSTANCE-IP-ADDRESS

Replace INSTANCE-IP-ADDRESS with your instance's IP address, which you can get from the Cloud dashboard .

The command will output, for example:

Starting Nmap 7.80 ( https://nmap.org ) at 2023-01-11 13:22 PST
Nmap scan report for 129.159.46.35
Host is up (0.041s latency).
Not shown: 999 filtered ports
PORT   STATE SERVICE
22/tcp open  ssh

Nmap done: 1 IP address (1 host up) scanned in 6.42 seconds

In the above example, TCP port 22 (SSH) is publicly accessible.

Note

If nmap doesn’t show TCP/22 (SSH) or any other ports open, your:

  • Instance might be terminated. Check the GPU Instances dashboard to find out.
  • Firewall rules might be blocking incoming connections to your instance.

Note

nmap -Pn INSTANCE-IP-ADDRESS only scans the 1,000 most common TCP ports.

How do I close my account?#

To close your Lambda GPU Cloud account:

  1. Back up all of your data on your instances as well as in your persistent storage file systems.

    !!! tip

    You can use rsync to back up your data.

  2. Terminate all of your instances from the Cloud dashboard or using the Cloud API.

  3. Delete all of your persistent storage file systems.

  4. In the Cloud dashboard, under Settings , click Close my account. Carefully read the warning in the dialog box that appears. To proceed with closing your account, type in close account, then click Close account.