HPE at SC25
November 16–21, 2025 | St. Louis, Missouri
Join HPE, world leader in supercomputing, at the International Conference for High Performance Computing, Networking, Storage, and Analysis, where innovation meets cutting-edge technology.
Unlock innovation and discovery in the AI era with HPE supercomputing
Unlock innovation, discovery, and new insights in the AI era with HPE supercomputing solutions and expertise. We’ll have our executives and subject matter experts onsite to discuss all your AI, HPC, and supercomputing needs.
- Interactive demos
- Hardware demos
Interactive demos at SC25
Sovereign AI and AI factory at scale
Organizations aim to maximize returns from investments in "AI Factory" GPU clusters for AI model training, tuning, inferencing, and mixed HPC workloads. This requires equitable, secure, and performant resource access, with self-service options for agility and tight cost monitoring. HPE offers an adaptable full-stack AI platform with control plane and workflow automations, enabling IT managers and service administrators to implement and scale these processes as AI workloads grow.
Data Center Networking for AI Training and Inference
HPE Juniper Networking portfolio for AI Data Centers includes switching, routing, security, and management & automation. The solution delivers open, secure, reliable, high-performance networks for AI training, inference, and AIaaS, offering design flexibility and simplified management and operations. Juniper was named a Leader in the 2025 Gartner® Magic Quadrant™ for Data Center Switching and ranked #2 for AI Ethernet Fabric Buildout in the 2025 Gartner® Critical Capabilities.
How to Build a Quantum Supercomputer
Scaling quantum computing (QC) for practical use, HPE Labs introduces a hardware-agnostic hybrid framework that tightly integrates HPC and QC across many layers of the stack. Scaling QC from hundreds to millions of qubits will require distributing workloads efficiently across multiple quantum processors (QPUs). The hybrid HPC-QC framework uses adaptive circuit knitting algorithms and large-scale HPC execution to enable efficient subcircuit identification and simulation, opening the door to practical quantum computing applications of the future.
HPE Cray Supercomputing Storage
Our purpose-built HPE Cray Storage systems are engineered to deliver intensive performance in the most-cost effective way, freeing up budget for investments in compute and science.
Direct Liquid Cooling at Scale: HPE Cray Supercomputing direct liquid cooling
As IT systems expand to meet the requirements of growing data and AI workloads, traditional air cooling will soon be unable to support the power needs of CPUs and GPUs drawing over 500 watts. Hear how HPE expertise in liquid cooling and our leadership portfolio span from the edge to Exascale—as a culmination of over five decades of experience and innovation. Explore our HPE industry-first 100% fanless liquid cooling systems architecture, it can help you achieve your performance, energy efficiency, and sustainability goals.
HPC High Performance Computing Software Portfolio including Exploratory Data Analysis at Scale: Arkouda Software
Enhance the user experience and performance of your HPC and AI workloads with supercomputing software. From system management, user services, and comprehensive development environments to our performance cluster manager, this comprehensive software toolkit will help you unleash the full potential of your HPC architecture.
Arkouda: As datasets grow beyond the capabilities of individual machines, data scientists require tools that leverage existing Python expertise while enabling scalable analysis. Hear how Arkouda, a Python package for exploratory data analysis at scale, offers a solution. Data scientists can leverage familiar Jupyter notebooks running on their laptops to analyze massive datasets, residing on cloud platforms, supercomputers, or other distributed machines. Arkouda is a generalized HPC framework for rapidly developing performant, scalable Python packages.
Hardware demos at SC25
HPE Slingshot Interconnect 400
Is your HPC networking solution able to meet your converged workload needs for today and tomorrow? See the HPE Slingshot Interconnect 400 that's now available with 400 Gbps and also as a TOR switch with PCIe NICs, ready to power the world's exascale systems. HPE Slingshot provides a modern, high-performance interconnect that delivers high bandwidth and low latency for HPC, ML, and analytics applications by bringing together the specialized requirements of HPC-optimized fabrics with the ubiquity of Ethernet.
HPE ProLiant Compute DL384 Gen12
Speed-time-to-value with HPE ProLiant Compute, take a tour of our AI optimized systems designed for edge and data center deployments—HPE ProLiant Compute DL145, DL380a, and DL384. Unlock more value from your data and accelerate AI outcomes that deliver insights and innovation.
HPE Compute Scale-up Server 3200
Scale-up compute solutions with the HPE Compute Scale-up Server 3200, the building block with in-memory compute for your business- and mission-critical solutions. Experience new ways to leverage the powerful capabilities for your most demanding enterprise applications - SAP HANA, Oracle database and the preview of SQL Server 2025, now with hybrid AI vector capabilities.
HPE Cray Supercomputing Storage Systems E2000
HPE Cray Supercomputing Storage Systems E2000 is the successor to Cray ClusterStor E1000 Storage Systems that have become the de-facto external parallel file system standard for exascale, pre-exascale, and national AI supercomputers. Learn why the E2000 systems are the perfect choice for supercomputers and large compute clusters running modeling and simulation, artificial intelligence, and high-performance data analytics applications in integrated workflows on one system.
HPE Coolant Distribution Unit
The HPE Cray Coolant Distribution Unit (CDU) is a liquid-to-liquid heat exchanger used to remove heat from HPE Cray Supercomputing EX4000 cabinets or HPE enterprise racks. It circulates heat transfer liquid in a secondary loop to cold sinks, transferring heat to the facility’s primary loop. Rated for 1.6 MW cooling, one “in-row” CDU supports up to four EX cabinets. It prevents condensation by keeping the secondary loop above the room’s dew point and offers top/bottom-feed plumbing options managed by a user-controlled system.
HPE Cray Supercomputing EX4000 and Coolant Distribution Unit
Experience the HPE Cray Supercomputing EX4000—a 100% liquid-cooled AI supercomputer using the HPE Coolant Distribution Unit that's capable of saving save the energy to power 896 fans in 14 enterprise racks when compared with conventional AI infrastructure. Learn how HPE has partnered with the U.S. Department of Energy, and AMD to help Lawrence Livermore National Laboratory to develop El Capitan, the world fastest, verified supercomputer that's powered by the HPE Cray Supercomputing EX4000.
NVIDIA Blackwell-based AI server: NVIDIA GB300 NVL72 by HPE
Unleash unprecedented AI power faster. Get massive real-time AI training and inferencing with expertise and support from HPE. See the future of AI supercomputing in a single rack, powering large AI models in one memory space with the latest NVIDIA technology. Discover five decades of innovation in liquid cooling and expertise. The server features 36 Grace CPUs and 72 Blackwell Ultra GPUs interconnected via NVIDIA NVLink™ in a rack-scale design, with up to 20 TB HBM3e GPU memory (72x288 GB).
HPE ProLiant Compute XD685 (NVIDIA)
The HPE ProLiant Compute XD685 server accelerated by NVIDIA GPUs is optimized for AI workloads like LLM and multimodal model training. It’s built on an HPE-designed modular chassis that speeds time to market, offers flexibility of GPUs and cooling methods, and optimizes rack density with up to eight nodes per industry-standard rack.
Unlock the power of large-scale AI with next-gen AI systems
This demo features HPE Compute XD690, optimized for service providers and large model builders. Featuring eight NVIDIA Blackwell B300 GPUs in a 10U air-cooled chassis, this powerhouse delivers breakthrough performance for building and training large AI models. HPE Services accelerate deployment, enabling you to get to market faster. Partner with HPE and NVIDIA to leverage the latest AI innovations and scale your AI capabilities globally.
HPE Cray Supercomputing EX254n Accelerator Blade
The HPE Cray Supercomputing EX254n Accelerator Blade with the NVIDIA GH200 Grace Hopper™ Superchip provides up to 2X performance for scientific computing, graph neural network (GNN) training, and inference applications over the prior generation, and provides unmatched NVIDIA tensor core GPU density as it packs eight NVIDIA Grace Hopper™ Superchips into a 1-rack-unit form factor. Join us to learn how HPE 100% liquid cooling technology makes that industry-leading density possible.
HPE Cray Supercomputing EX255a Accelerator Blade
The HPE Cray Supercomputing EX255a Accelerator Blade provides unmatched AMD APU density as it packs eight AMD MI300a APUs into a 1-rack-unit form factor. Explore how HPE 100% liquid cooling technology makes that industry-leading density possible when this accelerator blade is installed within the compute chassis of either the HPE Cray Supercomputing EX4000 or HPE Cray Supercomputing EX2500 systems.
HPE ProLiant Compute XD685 (AMD)
HPE ProLiant Compute XD685 is designed for complex AI model training tasks and powered by eight AMD Instinct MI355X accelerators. The new HPE system is optimized to quickly deploy high-performing, secure and energy-efficient AI clusters for use in large language model training, natural language processing, and multi-modal training.
HPE ProLiant Compute XD230
The HPE ProLiant Compute XD230 server is purpose-built for HPC workloads, delivering exceptional performance and efficiency. Powered by 2x Intel® Xeon® 6 processors with P-cores and advanced MRDIMMs, it achieves up to 8800 MT/s memory speed. Optional direct liquid cooling enhances sustainability and performance. HPE iLO and Silicon Root of Trust ensure security, while HPE Performance Cluster Manager enables comprehensive system monitoring and management.
HPE ProLiant Compute DL380a Gen12
The new HPE ProLiant Compute DL380a Gen12 server is optimized for AI with an ultra-scalable architecture delivering breakthrough performance, industry leading security innovation, and intuitive cloud operating experience: 4U 2P rack form factor, Intel® Xeon® 6 Scalable Processors, Up to 8 double-wide GPUs and Faster, efficient DDR5 memory and high speed PCIe Gen5 I/O bandwidth. Protect your infrastructure, workloads and data with a trusted edge-to-cloud security architecture. HPE’s compute core is hardened through a proven, zero-trust approach to security to minimize threats to hardware and risks from third-party software.
Experience Next-Gen HPC Networking with Juniper and University of Texas at Dallas at SC25
Join us at the University of Texas at Dallas (UTD) booth 4132 during SC25 in St. Louis, MO this November, where we’re featuring a high-performance computing demo in collaboration with Juniper Networks. Discover how Next-Gen 400G Ethernet stands up against traditional HPC networking solutions—delivering powerful performance and simplified operations.
- Live Demo: 400G Ethernet in Action
- Showing: Ethernet Performance for HPC Fabrics
- Discover: Real-world Benefits in Cost and Scalability
- Learn: Why 400G Ethernet is the Future of Research and HPC Networking
- Talk with Experts from Juniper Networks and UTD
Stop by the UTD booth and see why Ethernet is redefining what's possible in high-performance computing.
HPE Sessions at SC25
- Sunday, November 16
- Monday, November 17
- Thursday, November 20
- Friday, November 21
| 9:00 a.m.–3:00 p.m. CST Room 230 | The 8th Annual Parallel Applications Workshop, Alternatives To MPI+X (PAW-ATM 2025) |
| 2:00–5:30 p.m. CST Room 264 | Sustainable Supercomputing |
| 2:00–5:30 p.m. CST Room 274 | MEMO’25: International Workshop on Memory System, Management and Optimization |
| 2:00–5:30 p.m. CST Room 242 | High Performance Python for Science at Scale |
| 4:00–4:45 p.m. CST Room 130 | Speed, Smarts, and Science: Next-Gen HPC Networking at UT Dallas Brian Dourty, The University of Texas at Dallas |
| 9:00 a.m.–12:30 p.m. CST Room 261 | Seventh International Workshop on Emerging Parallel Distributed Runtime Systems and Middleware |
Celebrate with HPE at our featured receptions.
Champagne Toast
Monday, November 17, 2025, from 7:00–9:00 p.m.
Join the HPE executive team at HPE booth #3704 to celebrate our great year of accomplishments. This reception is an excellent opportunity to engage with peers, HPE executives, and subject matter experts in a relaxed environment while enjoying a great glass of champagne.
SC25 Industry Reception hosted by HPE and AMD
Tuesday, November 18, 2025, from 7:30–10:30 p.m.
Ballpark Village brings together premier dining, sports, entertainment, and nightlife in downtown St. Louis. Enjoy a special evening with colleagues and friends in a dynamic space perfect for enjoying food, spirits, and entertainment. All you need is your conference badge.
Platinum Sponsor of Women in HPC
Women in HPC is the only international organization working to improve equity, diversity, and inclusion in High Performance Computing. And at HPE, we know that our greatest strengths come from the people who make up our team. So, for us, diversity, equity, and inclusion are ingrained in who we are today because we know it’s what will get us to tomorrow. That’s why HPE is proud to be Women in HPC’s Platinum Sponsor at SC25. They have a full week of activities planned, and we’re looking forward to partnering with them once again.