TotalEnergies and Cerebras: Accelerating into a Multi-Energy Future (Updated) - Cerebras
Mar 11, 2025

Summary
This post details TotalEnergies' adoption of the Cerebras CS-2 system to accelerate multi-energy research, specifically seismic wave propagation modeling using stencil-based finite difference algorithms. The CS-2 system demonstrated over 200x performance improvement compared to a modern GPU. The post highlights the Cerebras architecture's advantage in memory-bound and communication-bound HPC workloads due to its wafer-scale integration, placing all processors on a single chip. It also mentions the Cerebras Software Development Kit (SDK) and Cerebras Software Language (CSL) as tools for developing custom kernels and leveraging wafer-scale computing.
More from the Wafer-Scale Field Equation Modeling trail
- Real-Time Computational Physics with Wafer-Scale Processing [updated]
- Beyond AI for Wafer Scale Compute: Setting Records in Computational Fluid Dynamics - Cerebras
- TotalEnergies and Cerebras Create Massively Scalable Stencil Algorithm - Cerebras
- Sandia Awards Major Contract to Cerebras Systems - Cerebras
- Cerebras Breaks Exascale Record for Molecular Dynamics Simulations - Cerebras