Overview
As the global energy industry accelerates toward AI-driven exploration, simulation, and energy optimization, TotalEnergies is making a major leap forward with the development of Pangea 5 — its next-generation high-performance supercomputer designed to dramatically expand seismic imaging, AI modeling, and energy research capabilities.
Built in collaboration with Dell Technologies, NVIDIA, and DDN, Pangea 5 will deliver an estimated 75 petaflops of compute performance when commissioned in 2027 at TotalEnergies’ Jean Féger Scientific and Technical Center (CSTJF) in Pau, France.
At the core of this system is DDN’s AI and HPC data intelligence platform — enabling the movement and management of massive, complex datasets at the speed required for modern AI and scientific discovery.
The Challenge
Energy exploration and AI-driven modeling require extraordinary computational scale and data throughput.
For TotalEnergies, the next generation of seismic engineering and subsurface imaging demanded significantly greater performance to:
- Improve the accuracy of subsurface geological imaging
- Accelerate exploration workflows and simulation cycles
- Support advanced AI and machine learning research initiatives
- Optimize integrated power and energy models
- Reduce operational and environmental costs
- Increase energy efficiency across infrastructure operations
As AI adoption expands across scientific and industrial workloads, the company needed infrastructure capable of supporting both traditional HPC and emerging AI pipelines without creating storage bottlenecks that could limit GPU and compute efficiency.
The Solution
DDN is delivering the high-performance data layer powering Pangea 5 through its EXAScaler platform with a hybrid flash and HDD architecture optimized for AI and HPC environments.
The deployment enables TotalEnergies to efficiently move, store, and manage enormous datasets required for seismic processing, AI model development, and advanced simulation workloads.
Key Infrastructure Components
- DDN EXAScaler high-performance parallel storage
- Hybrid flash + HDD architecture optimized for scale and throughput
- NVIDIA accelerated computing and InfiniBand networking
- Dell Technologies compute and rack infrastructure
- AI-optimized data movement and orchestration capabilities
By eliminating storage bottlenecks and maximizing data throughput, DDN helps ensure that Pangea 5 can fully leverage its accelerated compute environment for both AI and HPC applications.
Results & Impact
6X Increase in Compute Power
Pangea 5 is expected to multiply TotalEnergies’ computing power by six compared to previous generations, enabling faster processing and larger-scale simulations dramatically.
Enhanced Seismic Imaging Accuracy
The system will support advanced seismic engineering techniques that improve subsurface imaging precision, comparable to producing ultra-high-resolution “ultrasound scans” of underground geological formations.
Accelerated AI and Scientific Discovery
DDN’s high-performance data platform enables faster AI training, simulation processing, and analysis workflows critical to:
- Exploration optimization
- Integrated power modeling
- AI-assisted energy research
- Scientific R&D initiatives
Improved Energy Efficiency
Pangea 5 is also designed with sustainability in mind:
- ~40% lower energy consumption at equivalent performance levels
- Cooling system energy usage reduced by 5X
- Waste heat recovery used to help heat on-site facilities
Why DDN
Modern AI factories and supercomputing environments are no longer constrained solely by compute. Data movement, throughput, and orchestration increasingly determine overall system efficiency and time-to-insight.
DDN’s AI data intelligence platform was selected to power Pangea 5 because of its ability to:
- Sustain extreme-scale throughput for AI and HPC workloads
- Eliminate data bottlenecks across accelerated infrastructure
- Support hybrid AI + simulation environments
- Scale efficiently for next-generation scientific computing
- Deliver the performance density required for modern AI infrastructure
As AI transforms industries ranging from energy and climate science to autonomous systems and financial modeling, DDN continues to serve as the data intelligence foundation behind many of the world’s most advanced AI and HPC deployments.
About Pangea
TotalEnergies has leveraged the Pangea supercomputing platform for more than a decade to support advanced energy exploration and scientific research.
| System | Year | Performance |
| Pangea 1 | 2013 | 2.3 petaflops |
| Pangea 2 | 2016 | 6.7 petaflops |
| Pangea 3 | 2019 | 31.7 petaflops |
| Pangea 4 | 2024 | Hybrid cloud architecture |
| Pangea 5 | 2027 | ~75 petaflops (expected) |
Pangea 5 represents the company’s most ambitious investment yet in AI-powered energy innovation and scientific computing.
About TotalEnergies
TotalEnergies is a global integrated energy company that produces and markets energies: oil and biofuels, natural gas, biogas and low-carbon hydrogen, renewables and electricity. Our more than 100,000 employees are committed to provide as many people as possible with energy that is more reliable, more affordable and more sustainable. Active in about 120 countries, TotalEnergies places sustainability at the heart of its strategy, its projects and its operations.