VCU High Performance Research Computing Core Facility
Introducing Preetam Ghosh, Ph.D., Interim Director of VCU’s High Performance Research Computing Core
The HPRC is excited to announce Dr. Preetam Ghosh’s appointment as the interim director of VCU’s High Performance Research Computing Core. A distinguished researcher and educator, Dr. Ghosh brings a wealth of expertise and leadership to the HPRC, where he will guide the core’s efforts to support advanced research across the university.
Dr. Ghosh serves as the interim chair and professor in the Department of Computer Science in VCU’s College of Engineering. His research focuses on complex networks and systems biology, leveraging advanced techniques in modeling and simulation, applied artificial intelligence and network science to address societies’ grand challenges in the biomedical sciences.
With a career spanning academia and industry, the HPRC is poised to further its mission of empowering groundbreaking research through advanced computational capabilities and interdisciplinary collaboration under Dr. Ghosh’s leadership.
About the core
The High Performance Research Computing (HPRC) core facility provides the VCU community with the high performance computing resources for research, teaching and outreach activities.
The primary mission of the HPRC core facility is to advance scientific research at VCU in areas that benefit from large-scale computation. The HPRC supports the One VCU Research Strategic Priorities Plan and the recalibration of Quest 2028: One VCU Together We Transform by providing supercomputing clusters, large-scale storage systems, specialized software and technical expertise in research computing. The HPRC supports research computing in mathematics and computational sciences, physical and chemical sciences, engineering, life sciences, medicine and arts and humanities. Preetam Ghosh, Ph.D., interim chair and professor in VCU’s College of Engineering Department of Computer Science is the facility’s faculty director and Mike Davis is the technical director.
The HPRC operates two main clusters: Athena and Apollo, which provide service for different types of small/large, serial/parallel, CPU/GPU applications. The Apollo cluster is the only computing cluster certified at VCU to operate with CAT I data.
Frequently requested services include: parallel computing, GPU computing, large-scale storage, compilers, DNA/RNA/proteomics pipelines, computational fluids dynamics, quantum chemistry, and Artificial Intelligence.
Athena cluster
The primary cluster for large scale parallel computing.
- 114 compute nodes
- 9500 CPU cores
- 32 GPUs
- 63.5 TB RAM
- 3.1 PB of lustre filesystem
- High-speed InfiniBand architecture up to 200 Gb/s
Each node ranges:
- 28 – 238 cores
- 128 GB – 1.5 TB of RAM
- PU nodes include the NVIDIA V100, A100 and H100 GPUs
Apollo cluster
Supports research using data that complies with federal security and privacy requirements (CAT I data).
- 18 compute nodes
- 604 CPUs
- 2 Nvidia P100 GPUs
- 5.9 TB RAM
- 3 PB of Lustre high performance parallel file system storage
The Apollo system employs a security model that requires all access via VPN and exists on a separate virtual and physical network from other university and HPRC resources.
HPRC cluster upgrade
Because of the increased computing capacity, larger memory per node, and newer, faster CPUs, HPRC users will now be able to:
- run more jobs simultaneously
- run jobs 10x larger
- speed up the execution of jobs
- have less wait times in the queue
HPRC hardware upgrade
- 20 nodes: 128 CPU cores, 1.5TB RAM large memory machines
- 1 node: 4x H100 GPU, 64 core GPU server with 1TB RAM
- 3 switches: 200Gbit HDR infiniband switches and cables to upgrade connections
The HPRC anticipates a phased installation of these systems over the coming weeks, prioritizing large memory systems. Due to limited physical space in the faculty, HPRC staff must first remove older nodes to create space for new hardware.
These targeted upgrades will allow the HPRC to accommodate additional, larger jobs that align with VCU’s continuously expanding research enterprise.
HPRC 2024 investments and successes
In 2024, the HPRC increased its capacity to support AI and machine learning research through GPU equipment investments. With these investments, GPU computing hours surged by 90% from 76,431 in 2023 to 145,696 in 2024. These resources are primarily utilized by researchers in Computer Science and Mechanical Engineering for their AI/ML-driven projects.
A shining example of the transformative impact of the HPRC is the success of Ka Un Lao, Ph.D., an assistant professor in the Department of Chemistry in the College of Humanities and Sciences. Dr. Lao’s research in theoretical and computational chemistry has been significantly bolstered by the HPRC’s cutting-edge computational resources, advanced infrastructure, and the expertise of its dedicated team. The HPRC has provided the computational power and technical support needed to tackle highly complex problems, enabling Dr. Lao to develop innovative approaches that have led to groundbreaking discoveries.
With the support of the HPRC Core, Dr. Lao became the first faculty member at VCU to receive both the prestigious American Chemical Society Computers in Chemistry Division Junior Faculty Award and the NSF CAREER award from the Chemical Theory, Models and Computational Methods Program. These accomplishments underscore the critical role the HPRC plays in advancing his work, which has the potential to drive significant progress in sustainable energy solutions, molecular modeling, machine learning and other key areas of research. The HPRC continues to be an invaluable resource, fostering innovation and elevating the research capabilities of faculty at VCU
Why should VCU researchers use the HPRC?
- Cost-Effective and Convenient: The HPRC is centrally supported, eliminating the need for researchers to manage direct costs for hardware or maintenance for their research.
- Robust Resources: The HPRC boasts a substantial inventory of high-performance computing resources, providing researchers with the necessary infrastructure to conduct a wide range of research projects.
- Expert Support: Dedicated staff at the HPRC are readily available to assist researchers with their computational needs, providing guidance on resource allocation, troubleshooting and optimizing research workflows.
- Ensured Compliance and Safety: The HPRC ensures adherence to all relevant regulations and safety standards, providing a secure and compliant environment for researchers to conduct their work.
By the numbers…
- 700+ VCU users of the HPRC
- 2 high performance computing clusters delivering more than 40 TB of RAM.
- 36 million computing hours generated by the HPRC over the past year
- 8-out-of-10 VCU research projects produce data that can be analyzed using the HPRC.
Meet the team
Preetam Ghosh, Ph.D.
Interim director
[email protected]
Carlisle Childress
Operating systems, security, hardware, storage
[email protected]
John Layne
Programming, parallelism, system utilities
[email protected]
Mike Davis
Technical director
[email protected]
Brad Freeman
Operating systems, hardware accounts
[email protected]
Romano Woodfolk
Operating systems, security, compliance
[email protected]