Information Technology Services

Home » Research » HPC

High Performance Computing Support

Information Technology Services (ITS) has entered a partnership with NC State to provide UNCG faculty, staff and students with access to the "Henry 2" High Performance Computing (HPC) Linux cluster. The Henry 2 cluster is comprised of ~1,000 computer nodes, many of which have multi-core processors, providing computing power that far surpasses that of a desktop computer.

For more detail regarding the Henry 2 Linux cluster architecture and available software, see:

HPC Linux Cluster Account Requests

UNCG faculty, staff, and students may request an account for "Henry 2" at no charge. To request an account, please contact 6-TECH at (336) 256-TECH (8324) or 6-TECH@uncg.edu - or visit 6-TECH Online to open a request. Students requesting an account should submit written proof of faculty sponsorship (email is fine).

Account requests are typically processed in 1-2 weeks. Once your account is established, you may request support by contacting 6-TECH.

HPC Linux Cluster Software Requests

UNCG faculty and staff may request that appropriately licensed software be installed on the "Henry 2" cluster. Before making a request, please review the current list of software on Henry 2 to make sure the software is not already available. Requests may be submitted via the Application Request form.

Note: Select Henry 2 – HPC Linux Cluster as the "Environment."

ITS will then review the request for technical compatibility with the Henry 2 environment and for appropriate licensing. If technical compatibility and licensing requirements are met, software installation requests are typically completed within 2 weeks.

HPC Support and Consultation

ITS and NC State share a position for technical support of faculty, research staff, and students (with a faculty sponsor) who use the HPC Linux cluster Henry 2.

For FY 11-12 only, ITS will cover the cost of technical support to UNCG clients at no charge back to departments. Depending on the level of interest and use, the support and training partnership may continue beyond 2011-12 with a cost-sharing arrangement between interested UNCG departments/researchers and ITS.

To request support, please contact 6-TECH at (336) 256-TECH (8324) or 6-TECH@uncg.edu - or visit 6-TECH Online.

High Performance/Parallel Computing Training Opportunities at UNCG

Pending confirmation of sufficient interest, ITS will host on-site high performance/parallel computing training for faculty, research staff, and students. Other training sessions may be scheduled via telephone/video and web conferencing.

An Introduction to High Performance/Parallel Computing is available onsite at UNCG. This workshop provides an introduction to the Henry 2 Linux cluster, MPI (Message Passing Interface), and offers lab time for individual attention with specific research problems (see below for more details).

Currently scheduled workshop dates are listed under High Performance/Parallel Computing on the UNCG Workshops and Event Calendar. To request alternative training dates for your department or group (minimum 5 participants) contact 6-TECH.

Prior to participating in a workshop, those do not have an account on Henry 2 should contact 6-TECH as soon as possible to request an account (256-8324 or 6-TECH@uncg.edu).

Workshop description - Introduction to High Performance/Parallel Computing:

Graduate students, postdocs, faculty and staff who have an interest in using parallel computation in research or instructional projects are invited. Before the workshop starts, faculty, staff and students who do not already have an account are encouraged to request one by contacting 6-TECH (256-8324 or 6-TECH@uncg.edu) or visiting 6-TECH Online to open a request.

Through a partnership with NC State, UNCG researchers have access to several thousand processors available for high performance computing. This short course introduces the use of the machines, starting with how to log on and submit jobs. The course introduces the use of MPI (Message Passing Interface), the standard library for message passing parallel computation. Calls to MPI are embedded in Fortran, C, or C++ codes, enabling many processors to work together to accomplish a parallel computation. Lab time will be available for individual assistance with parallel computing questions.

Topics include:

  • How to log into the HPC machines and submit jobs. Why to use parallel computation. Some simple MPI commands and example programs.
  • MPI Collective communications. These can be simple and efficient. Considerations in efficient parallel computation.
  • Overlapping computation and communication.

Links to course material:

Please spread the word about this opportunity for HPC training!

Related Links

Qualtrics - Web Survey Software