Frequently Asked Questions
If you have questions that are not included in this page or in rest of the documentation, please email hpc@umass.edu.
How do I connect to the Unity cluster and begin using it?
There are multiple ways to connect to the Unity cluster:
- Unity OnDemand is the simplest and most convenient method of connecting because it allows you to login through your browser.
- SSH Connection is the most traditional method of connecting to the cluster, especially for experienced Command Line Interface (CLI) users. SSH connection allows you to connect to Unity in your terminal.
- Visual Studio Code Desktop allows you to connect to Unity using the Remote-SSH feature, which is useful for users that prefer to use VS Code as their editor.
- PuTTY is a convenient way to connect to the cluster that is available to Windows users. See the SSH Connection guide for instructions on how to connect with PuTTY.
When I connect over SSH I get a message saying permission denied (public key)
This can be due one of these common reasons:
- You have not provided your private key while connecting. Use
ssh -i <private_key_location> <user>@unity.rc.umass.edu - You are not assigned to at least one PI group. We require at least one PI to endorse your account before you can use the cluster. Request to join a PI on the My PIs page.
- You have not added a public key to your account on Unity yet. You can do this on the Account Settings page. For more information about public and private keys, see our guide on SSH Connection.
- Your login shell is invalid. In Account Settings, try “/bin/bash” or “/bin/zsh”.
- You are a PI, and you are trying to use your PI group name to log in. Your login username should not start with
pi_.
Where can I find software to use on the cluster?
Most of our software is package installed and is available by default.
Unity OnDemand provides JupyterLab, Matlab, RStudio, Mathematica, and a XFCE, a general purpose interactive desktop environment.
Non standard and version specific software are available as modules.
- To print all available modules, use the command
module av. - To filter the available modules, use
module av <name>. - To load a module and gain access to its binaries (executables), use
module load <name>.
To install additional software, see our guide on conda environments.
I’m looking for xyz software; could you install it?
Most software that is requested is free for use. If this is the case we will install it for you, just send us an email at hpc@umass.edu titled “software request: <name>”. If the software you want is licensed, we may be able to help since the campus often has site-wide licenses for many applications.
Can I run containers on Unity?
Yes! We support Apptainer (formerly Singularity) containers, which are fully compatible with Docker images. Run module load apptainer/latest to access it.
How much storage do I get on Unity and is it backed up?
Refer to storage information here.
We do not provide backup solutions by default.
We take snapshots of /home/ and /work/ every day at 1AM, but delete them after two days.
When I try to queue a job, I get denied for MaxCpuPerAccount.
Resource limits are set per lab. Currently, they are 1000 concurrent CPUs and 64 concurrent GPUs. This allocation is shared across your entire PI group.
I’m a PI and I would like to purchase hardware to buy-in to Unity.
Great! Send us an email and we’ll be happy to help. We are very flexible when it comes to the needs of research labs.
I’m part of multiple PI groups; how can I make sure I’m doing my work under a certain group and not a different one?
Use --account=pi_... to change an association for that job.