ODU maintain a number of high-performance computing (HPC) clusters. An HPC cluster is a set of computers connected via high-performance network to allow them to operate as a single "supercomputer". An HPC cluster is a shared computing resource, which means that the CPUs, memory and storage on a cluster is used and shared by many users. Currently there are two HPC clusters, Wahab and Turing:
Cluster Num of CPU cores Num of GPUs Memory (aggregated) Scratch space Wahab ~6320 72 60 TB 350 TB Turing ~5600 36 34 TB 180 TB
More details can be found in our main website. Both Wahab and Turing clusters use SLURM job scheduler to manage workloads on their computing resources.
Before we dive into the important topics such as the SLURM Scheduler, environment management, and details about our clusters, we are going to cover some basic concepts and answer a few frequently asked questions.
Use your Monarch Identification and Authorization System account (MIDAS) to access High Performance Computing resources. If you are an ODU student/faculty/staff who do not have a MIDAS account yet, please visit here.
To request the activation of the HPC service on your account, please fill in the Request HPC Resources form on our main site. External (non-ODU) persons that need access to our HPC resources in collaboration with ODU faculty or staff will also fill out the same form and indicate their faculty sponsor.
When you need help, the Research Cloud Computing group is always here for you. Reach us by email at rcc@odu.edu .
Since 2020, we have been hosting a virtual office hour using Zoom at 11 AM every Wednesday, please refer to our new user message for detailed information. During this time we are happy to help with any questions or issues you might have.
New users are strongly encouraged to watch the Introduction to HPC webinar to familiarize themselves with our HPC environment and how to properly use the HPC resources.
There are three ways to remotely access our resources:
Secure Shell (SSH)
SSH is an encrypted network protocol to interact with a computer remotely. It enables users to remotely login to our servers and operates securely over the Internet. It also allows Secure File Transfer Protocol (SFTP) for file access to our cluster. Using SSH, you can connect to our resources from campus, your home, and anywhere else.
Open OnDemand
Wahab cluster features Open OnDemand, a web-based access to its rich HPC capabilities. Please refer to Open OnDemand documentation page to learn more. Many users who are not familiar with terminal interaction may find this interface more friendly.
Remote Desktop Protocol (RDP)
X11 is the foundation of UNIX graphical interfaces. It allows users to reach our resources graphically and remotely. X11 over Remote Desktop Protocol (RDP) for the Turing and Wahab clusters. While X11 forwarding is also supported over SSH, it is not recommended unless the system does not support RDP.
Here is a matrix of the connection methods supported by our systems:
Cluster Cluster Hostname SSH Support Open OnDemand X11 Support X11 over RDP Support File Transfer Support Wahab wahab.hpc.odu.edu yes yes yes yes yes Turing turing.hpc.odu.edu yes no yes yes yes
Please replace
CLUSTER_HOSTNAME
with the Cluster Hostname listed in the table when following the examples below.
Beginning June 6th, 2022, you are required to use DUO two factor authentication when accessing the Turing and Wahab clusters.
DUO two-factor authentication is required when connecting to the Wahab and Turing clusters from off campus or from ODU wireless network. Two-factor authentication will not be enforced when using the VPN or when connecting from on campus. DUO two factor can be enabled in MIDAS. For more information please see the MIDAS help page: https://www.odu.edu/ts/access/two-factor-authentication.
SSH is installed on most UNIX-like systems by default. You can use the following command in your terminal to connect to our resources:
ssh YOUR_MIDAS_USERNAME@CLUSTER_HOSTNAME
In case you cannot find the ssh
command, please try installing the following package with the package manager that ships with your OS.
openssh-client
openssh-clients
If connecting from off campus while not using the VPN, you will prompted for your 2-factor authentication information.
Please select the 2-factor method you would like to use. If using the standard DUO client on your mobile device please select option 1. for "Duo Push to XXX-XXX-XXXX".
On UNIX-like desktop systems, GUI programs running on the clster can have their displays forwarded to your local desktop. This is very easy to do and requires no configuration:
ssh -C -X YOUR_MIDAS_USERNAME@CLUSTER_HOSTNAME
# -X enables X11 display forwarding
# -C compresses the data to enhance graphical performance
Direct X11 display forwarding can have very slow performance for many applications, however. Also, when the connection is disrupted, the GUI program will also terminate. If your applications behave very sluggish, you may want to consider two other options: X11 over RDP (see below), or web-based virtual desktop on Open OnDemand.
For Mac users:
If you have never used X on your system before, this command will install XQuartz (the X11 implementation on OSX) for you. When the dialog prompts for the installation, please allow it to complete.It can also be downloaded from their website.
Older versions of Windows do not come with a SSH client but there are many great, free options available. We list a few that are popular among SSH users on Windows:
Update 2020: Recent versions of Windows 10 now includes the
ssh
command that can be called from the command-line interface (CMD and PowerShell). Please launch "CMD" on your Windows desktop, and try ifssh
command exists. If it does, then you will use the instruction for Linux/Mac.
Putty does not require an installation procedure. You can just download the stand-alone binary (64-bit Windows download link here) and double click it to execute. We recommend new users to use Putty for Windows SSH access. Putty can remember the host name and other SSH settings for you.
Step 1: Start Putty.
Step 2: Enter the cluster hostname (turing.hpc.odu.edu
or wahab.hpc.odu.edu
).
Step 3: If this is the first time you connect to the cluster, click YES on the dialog shown below to continue.
Step 4: Enter your credentials (MIDAS ID and password).
RDP stands for Remote Desktop Protocol, allowing you to access a remote GUI desktop as if it were your own desktop.
After logging in to the cluster with RDP, you will receive a DUO push notification on your primary device registered with DUO. After entering your username and password, you will not see any indication or message regarding the DUO push. Simply open your primary DUO device and approve the request to login to HPC. Your login will not be completed until you acknowledge the request using the DUO app on your primary device.
We suggest X11 over RDP for you since there is a built-in remote desktop connection application for every Windows installation. This approach requires minimal configuration.
Step 1: Start Remote Desktop Connection.
You can find it from searching in your Start menu.
Step 2: Enter the cluster information.
Step 3: Enter your credentials.
MobaXTerm is a fully-featured client software that supports not only SSH, but also X11 over SSH and RDP. We recommend Windows power users to take advantage of MobaXTerm.
By default, Mac does not come with a RDP client but you can download the Microsoft Remote Desktop Connection App from the App store.
The app enables you to connect to our resource over RDP.
Once installed, please perform the following steps to establish a connection to the cluster.
Step 1, Start your app and click the New button
You will see this dialog asking for your information. You don't have to do this every time.
Input your information and click the close button. There is no save button here.
Double click on the entry you just created and you will have your connection.
You have several choices. Usually, a remote desktop viewer comes with your distribution. If it is not available, you may want install one of following clients:
Since 2020, we have offered a web-based interface to Wahab HPC cluster via Open OnDemand, a project of Ohio Supercomputing Project. Point your browser to https://ondemand.wahab.hpc.odu.edu/ to get started. Please refer to our Open OnDemand documentation page to learn more.
From the OnDemand interface, you can access, manage, upload and download your files on HPC, monitor and submit jobs, launch interactive sessions such as Jupyter, RStudio and virtual desktop.
Q: Can I use any of the resources offered by the Research Computing Group?
A: Yes, if you are a valid Old Dominion University student or a member of the faculty or staff. Also yes, if you are conducting research and collaborating with ODU faculty or staff.
Q: Does the Research Computing Group offer software X or service Y?
A: If you can not find it in our documentation, you can always contact us and let us know what you want. We try the best to accommodate every user's request. Free and open-source software can usually be installed on our system; proprietary software will require a paid license before it can be installed by our system, to be used only by the authorized users.