Synopsys is adopting NVIDIA RTX PRO Servers, a new category of enterprise data centre infrastructure built on the NVIDIA RTX PRO 6000 Blackwell Server Edition GPU — to accelerate compute-intensive, Al-driven simulation across silicon, systems, and AI.
Boasting double the memory capacity of the previous generation, NVIDIA RTX Pro Servers accelerated by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs will enhance simulations, rendering, and data speeds, handling detailed models more efficiently than previous generations.
The RTX 6000 PRO Server is anticipated to support simulation workloads of exceptional scale, including Ansys Fluent and Ansys Lumerical FDTD models with billions of cells. The full list of Synopsys products running on RTX PRO Servers includes:
- Ansys Fluent fluid simulation software
- Ansys FreeFlow smoothed-particle hydrodynamics simulation software
- Ansys Lumerical FDTD 3D electromagnetic simulation software
- Ansys Perceive EM radio frequency channel & radar signature simulation software
- Ansys Rocky particle dynamics simulation software
- Ansys Speos integrated optical & lighting simulation software
Engineers face growing complexity and an increasingly rapid pace of innovation. Synopsys is at the forefront of applying AI across its electronic design automation (EDA) and Ansys simulation and analysis portfolio to address these challenges and maximise the productivity of engineering teams.
Delivering AI-powered engineering solutions requires cutting-edge AI infrastructure that can handle computationally intensive workloads from complex external aerodynamic analyses to large-scale wireless network and radar modelling. That’s why Synopsys is also at the forefront of applying AI infrastructure, including NVIDIA RTX PRO Servers, to cost-effectively speed the development and delivery of its next-generation products.
NVIDIA RTX Pro Servers’ features complement simulation and analysis workloads:
- Ray tracing power: featuring 188 RT cores for modelling light and electromagnetic wave behaviour, NVIDIA RTX PRO 6000 Blackwell GPUs enhance the efficiency of Perceive EM for radar signal simulation and Speos for optical light study — boosting simulation speed and reliability while assisting engineers in optimizing designs more effectively
- High memory & dandwidth: equipped with 96 GB of GDDR7 memory and a 512-bit bus, delivering ~1.79 TB/s bandwidth — NVIDIA RTX PRO 6000 Blackwell GPUs are well suited for simulation workloads that demand large computational domains and high-resolution field data
- With doubled memory bandwidth, Lumerical FDTD simulations could achieve speeds up to 2x faster than with the prior NVIDIA L40S generation of GPUs
- The doubled memory capacity allows Fluent models with up to twice the number of cells, supporting higher fidelity physics
