Help

Get quick access to help resources and information for high-performance computing (HPC) systems.

Support Contact Information

General HPC assistance,  requests, troubleshooting technical problems

Please include the project handle and username(s) in the email and any information that will help us identify and troubleshoot the issue.

HPC-Help@nrel.gov

General HPC assistance, requests, troubleshooting technical problems

This contact form also generates an email to HPC-Help@nrel.gov. Please include the project handle and username(s) in the email and any information that will help us identify and troubleshoot the issue. 

Contact us

HPC allocations

Email for questions about the allocation request itself and increases/decreases to AUs and/or storage. 

HPC-Help@nrel.gov 

Help Resources

Computational Sciences Tutorials team

Staff in the Computational Science Center host tutorials and workshops on computational science topics—such as visualization, cloud, and HPC—throughout the year. In Microsoft Teams, the Computational Sciences Tutorials public team is the hub for all such tutorials and workshops. 

Computational Sciences Tutorials 

HPC website

The website offers resources to get access to systems, basics on getting started and running jobs, accounts and allocation information, and our policies.

www.nrel.gov/hpc/

Virtual, real-time assistance

Bring your HPC-related questions or issues and discuss with technical staff and other users every other week on alternating Tuesdays and Thursdays. 

HPC office hours

GitHub repository

Find resources to improve the user experience for both novices and veterans. The repository contains a collection of code examples, executables, and utilities. The GitHub Pages site provides more verbose documentation and examples and contains documentation on auxiliary systems such as Swift and Vermilion.  Both the repository and GitHub pages are open to contributions from the user community. 

HPC GitHub Repository

HPC GitHub Pages 

HPC data and ESIF research data systems

Contact this email address to discuss your data science needs or questions with a data expert.   

data.experts@nrel.gov

Computational Sciences Tutorials Team and Channels

Staff in the Computational Science Center host tutorials and workshops on computational science topics—such as visualization, cloud, and HPC—throughout the year. In Microsoft Teams, the Computational Sciences Tutorials public team is the hub for all such tutorials and workshops.  If you want to receive notifications; seek out new or upcoming tutorials; and see past slide decks, repo links, and recordings, join the team.

To add or remove members from the Computational Sciences Tutorials team, contact Mo Nelson. If you have questions about the Computational Sciences Tutorials team, including how to access/use features within Teams, contact Jennifer Southerland.

We update our members annually from the project team list. Unfortunately, this may mean you are added to the team even after you have left. We do not have a way to cross-reference but will start a running list this year of people who ask to be removed. You can hide the team and turn off notifications as an alternative to leaving the team. 

For internal users (NREL), follow these instructions to join:

  1. In Teams, click on the "Teams" icon in far left nav.
  2. Click "Join or create a team" in the lower-left corner.
  3. In the "Search teams" field in the upper right, type "Computational" and hit return.
  4. Click Join.

As an external user, you will be able to view discussion board posts, resource files, our SharePoint Calendar, and related repo links (all tabs at the top under General). The SharePoint Calendar provides a month view of upcoming tutorials and links to join. If you miss the monthly announcements in our newsletters, you can access calendar events and find meeting information to join the tutorials in the Teams channel. Unfortunately, you will not be able to access recordings or the survey. If you want access to a recording, contact Emily Harrell.

To join the public team, first register for a free Office365 account or log in if you already have an account. Next, download Microsoft Authenticator or another authenticator application. The process is straightforward, and you will be prompted during each step.

If you do not accept the invite or do not wish to go through the process of joining the public team, you can rely on the monthly newsletters or visit the training page on the website for meeting information. 

If you have received an invite from the team: 

  1. You will receive a welcome email from the team owner with information about the team. Click on Accept. 
  2. If you have never created a Microsoft Office 365 account, you will be prompted to create one. If you already have a Microsoft Office 365 account, log in. 
  3. The first time you log in, you will be prompted to set up Microsoft Authenticator.
    • From your mobile device, download and install the app from the Apple Store (for iOS) or the Google Play Store (for Android) and open the app.
    • On your mobile device, you will be prompted to allow notifications. Select Allow.
    • On your mobile device, click OK on the screen for what information Microsoft gathers.
    • Click Skip on the "Add personal account" page.
    • Click Skip on the "Add non-Microsoft account" page.
    • Click Add Work Account on the "Add work account" page.
    • Click OK to allow access to the camera.

Going forward, anytime you login, you will get a prompt on your phone to authenticate. 

Frequently Asked Questions

Generally speaking, HPC infrastructure is coordinating many discrete units capable of independent computation to cooperate on portions of a task to complete far more computation in a given amount of time than any of the units could do individually. In other words, an HPC system is lots of individual computers working together.

HPC Operations and Information Technology Services (ITS) are separate groups with different responsibilities. ITS will handle issues with your workstation or any other digital device you are issued by NREL. HPC Operations will assist with issues regarding HPC systems. Note that your NREL HPC account is a different account from your ITS credentials that you use to login to your workstation, e-mail, and the many other IT services provided by the Service Desk.
Over the fiscal year, there is a given amount of time each computer in the HPC system(s) can be expected to be operational and capable of performing computation. HPC project allocations allocate a portion of the total assumed available computing time. The sum of all awarded project allocations' compute-time approximates the projected availability of the entire system. Project allocations are identified by a unique "handle" which doubles as a Linux account under which you submit HPC jobs related to the project to the job scheduler. Learn more about requesting an allocation.
OTP tokens are a means of two-factor authentication by combining a temporary (usually lasting 60 seconds) token to use along with your account password. Tokens are generated using the current time stamp and a secure hashing algorithm. Note that you only need an OTP to access systems outside the NREL firewall, namely if you are an external collaborator. NREL employees can be on-site or use a VPN to access HPC systems via the *.hpc.nrel.gov domain.
VPNs simulate being within a firewall (which is an aggressive filter on inbound network traffic) by encapsulating your traffic in a secure channel that funnels through the NREL network. While connected to a VPN, internal network domains such as *.hpc.nrel.gov can be accessed without secondary authentication (as the VPN itself counts as a secondary authentication). NREL employees may use the NREL VPN while external collaborators may use the NREL HPC VPN using their OTP token. This provides the convenience of not having to continually type in your current OTP token when accessing multiple systems in a single session.
This is the general term used for any task submitted to the HPC systems to be queued and wait for available resources to be executed. Jobs vary in how computationally intensive they are.
A node is a complete, independent system with its own operating system and resources, much like your laptop or desktop. HPC nodes are typically designed to fit snugly in tight volumes, but in principle you could convert several laptops into a cluster, and they would then be "nodes."
Login nodes are the immediate systems your session is opened on once you successfully authenticate. They serve as preparation systems to stage your user environment and launch jobs. These login nodes are shared resources, and because of that the HPC team employs a program called Arbiter2 to ensure that these resources aren't being used inappropriately (see 'What is proper NREL HPC login node etiquette' for more detail). Compute nodes are where your jobs get computed when submitted to the scheduler. You gain exclusive access to compute nodes that are executing your jobs, whereas there are often many users logged into the login nodes at any given time.

As mentioned above, login nodes are a shared resource, and are subject to process limiting based on usage. Each user is permitted up to 8 cores and 100GB of RAM at a time, after which the Arbiter monitoring software will begin moderating resource consumption, restricting further processes by the user until usage is reduced to acceptable limits. If you do computationally intensive work on these systems, it will unfairly occupy resources and make the system less responsive for other users. Please reserve your computationally intensive tasks (especially those that will fully utilize CPU cores) for jobs submitted to compute nodes. Offenders of login node abuse will be admonished accordingly. For more information please see our policy on what constitutes inappropriate use.

System time is a regularly occurring interval of time during which NREL HPC systems are taken offline for necessary patches, updates, software installations, and anything else to keep the systems useful, updated, and secure. You will not be able to access the system or submit jobs during system times. System times occur the first Monday every month. A reminder announcement is sent out prior to every system time detailing what changes will take place, and includes an estimate of how long the system time will be. You can check the system status page if you are ever unsure if an NREL HPC system is currently down for system time.
As you become familiar with navigating the HPC Linux systems you may come to prefer to use the same command-line interfaces locally on your workstation to keep your workflow consistent. There are many terminal emulators that can be used on Windows which provide the common Linux and macOS command-line interface. The official Linux command-line emulator for Windows is known as the Windows Subsystem for Linux. Other recommended terminal applications include Git Bash, Git for Windows, Cmder, and MSYS2. Note that PuTTY is not a terminal emulator, it is only an SSH client. The applications listed above implement an ssh command, which mirrors the functionality of PuTTY.
Stated briefly, the SSH protocol establishes an encrypted channel to share various kinds of network traffic. Not to be confused with the ssh terminal command or SSH clients which are applications that implement the SSH protocol in software to create secure connections to remote systems.
Good question! There may be hundreds of reasons why. Please contact HPC Support with a message containing as many relevant details as you can provide so we are more likely to be able to offer useful guidance (such as what software you're using, how you are submitting your job, what sort of data you are using, how you are setting up your software environment, etc.).

Share