NERSC Issues a Request for Proposals for 2026: Research on Quantum Information Science Can Use Up to 20,000 GPU Node Hours
QIS@Perlmutter
To use the potent Perlmutter supercomputer for quantum information science (QIS) studies, the National Energy Research Scientific Computing Centre (NERSC) has launched the QIS@Perlmutter 2026 Call for Proposals. With accepted ideas potentially getting up to 20,000 graphics processing unit (GPU) node hours, this program provides significant computational resources. The U.S. Department of Energy Office of Science’s mission computing facility, NERSC, is housed at Lawrence Berkeley National Laboratory (Berkeley Lab) and supports research in a variety of disciplines, including physics, chemistry, and materials sciences.
You can also read Oxigen Data Center With Qilimanjaro For Quantum Data Center
Accelerating QIS and Advancing Hybrid Computing
The QIS@Perlmutter program’s main objective is to develop quantum tools, methods, software, and algorithms to progress QIS. Enabling hybrid computing, which attempts to connect quantum systems with classical high-performance computing (HPC) to make advances in fields like materials science and chemistry, is a major focus.
Accepted research covers a wide range of important topics in quantum science. All facets of quantum information science are open to projects, including:
- Quantum modelling of chemical systems and materials.
- Quantum circuit compilation and simulation algorithms.
- Quantum computing error prevention methods.
- Creation of software for the stack of quantum computing.
- The creation and evaluation of hybrid quantum-classical codes.
- Studies examining the relationships between conventional HPC systems and quantum computing systems (or accelerators).
The software speeds up quantum science by using state-of-the-art classical computer capacity, particularly Perlmutter’s GPUs, to solve difficult issues.
Perlmutter’s High-Performance Resources and Software
The main platform for this allocation is the Perlmutter supercomputer, which has NVIDIA A100 GPUs. Up to 20,000 GPU node hours are available to selected candidates for the NERSC 2026 Allocation Year (AY). Crucially, there are four A100 GPUs in per Perlmutter GPU node.
A comprehensive collection of software and libraries necessary for contemporary quantum research will be available to researchers. These include the CUDA-QX and cuQuantum packages from NVIDIA, which make it easier to simulate tensor networks and state vectors on GPUs. Qiskit, PennyLane, Quimb, and ITensor are more pertinent software programs that are accessible.
Additionally, successful candidates will have the exceptional chance to collaborate with Nvidia and NERSC personnel. The goal of this partnership is to help awardees with workflow optimization, project-related problems, and efficient GPU use.
You can also read How Bosonic Quantum Processors Scaling Quantum Computers
Eligibility, Criteria, and Allocation Year Details
All applicants are welcome to apply, and prospective users are not need to be active NERSC users. Proposals are welcome from researchers in academia, industry, and national labs, especially those in computer science, physics, and materials.
Proposals are assessed based on particular utilization criteria in addition to their investigation of a quantum information science topic:
- Leveraging Perlmutter Features: Candidates must present a well-thought-out strategy for using the system’s key components, which include an all-flash scratch file system and more than 6,000 NVIDIA A100 GPUs.
- Preparing for Doudna Capabilities: Proposals should explain how they intend to make use of the future Doudna supercomputer’s additional features. These features include support for interactive workflows, a high-speed Quantum-X800 InfiniBand networking infrastructure, and the NVIDIA Vera-Rubin CPU-GPU platform.
- Societal and DOE Impact: Projects that show a significant benefit to the Department of Energy Office of Science’s present or future research goals or to society at large will be given priority consideration.
The allocation covers the period from January 16, 2027, to the NERSC 2026 Allocation Year (AY). Projects have to be finished by this date. Awardees must provide a final summary at the conclusion of the project and report progress on a bi-monthly basis.
You can also read MapLight Therapeutics News: Collaboration For CNS Therapies
Application and Open Science Requirements
Online applications are now being accepted, and they are examined on a rolling basis. All submissions must be received by March 1, 2026, to be considered.
All findings from projects that are awarded must be published in open scientific publications or presented in open forums since NERSC is a national user facility of the Department of Energy Office of Science that is devoted to open scientific research. Standard NERSC acceptable use policies must also be followed by every project. An “Institutional User Agreement” is necessary if the proposed study does not get federal support.
Ermal Rrapaj, Daan Camps, Neil Mehta, and Katherine Klymko are the QIS@Perlmutter team’s contacts. The initiative provides high-impact researchers with the computational power they need to propel advances in the quickly developing field of quantum information science.
You can also read Purification Light for Quantum Technology with Optical Noise
