“Synopsys has a proud history of empowering engineering teams to solve previously unsolvable challenges, and now we’re taking that to the next level by harnessing the power of AI and accelerated computing,” said Sassine Ghazi, president and CEO of Synopsys.
The company used the event to describe the performance potential of its EDA stack driven by NVIDIA’s GH200 Grace Hopper Superchip and unveiled new initiatives with NVIDIA to help maximise the capabilities of engineering teams working on technologies from silicon to automotive systems.
Synopsys is applying NVIDIA accelerated compute architectures, including the NVIDIA GH200 Grace Hopper Superchip, to achieve significant, projected runtime gains of up to 15X compared to current methods, across the full EDA stack spanning design, verification, simulation and manufacturing.
Among the announcements made:
- Synopsys VCS: Simulation and constraint solver engines will accelerate functional verification workloads on NVIDIA GPUs. Using advanced Fine-Grained Parallelism (FGP) technology and massively parallelised graph evaluation enables users will be able to find bugs earlier, faster and smarter.
- Synopsys Fusion Compiler: Enables accelerated hybrid CPU/GPU performance scaling across digital placement, providing massive parallelisation for compute-intensive exploration and optimisation.
- Synopsys PrimeSim: Accelerates SPICE simulation workloads by leveraging NVIDIA GPUs. The heterogeneous, accelerated compute architecture enables the simulation of challenging circuits to achieve signoff with SPICE-level accuracy, reducing runtimes from days to hours.
- Synopsys Proteus: Provides optical proximity correction software (OPC) for the acceleration of computational lithography, the largest workload in the semiconductor manufacturing process.
Synopsys OPC software running on the NVIDIA cuLitho software library dramatically accelerates computational lithography workloads compared to current CPU-based methods.
“Our collaboration with Synopsys on generative AI and digital twins is central to the future design, automation and manufacturing of chips,” said Jensen Huang, founder and CEO of NVIDIA. “Synopsys’ EDA suite, together with NVIDIA’s accelerated computing architecture, demonstrates the gains that can be achieved for these exceptionally demanding industrial workloads.”
In addition, Synopsys and TSMC are going to production with NVIDIA cuLitho, a computational lithography platform, to accelerate manufacturing and push the limits of physics for the next generation of advanced semiconductor chips.
Synopsys is extending its Synopsys.ai LLM-based capabilities, beginning with Synopsys.ai Copilot, to support NVIDIA AI and compute platforms, giving customers more flexibility to customise their data sets and enable air-gapped on-prem deployment. Synopsys.ai Copilot is a generative AI capability that is designed to help engineering teams accelerate time to market and address systemic complexity through the power of conversational intelligence.
Synopsys will utilise the NVIDIA AI Enterprise software platform, which includes NVIDIA NeMo framework, and NVIDIA NIM inference and NeMo Retriever microservices deployment containers.
Synopsys customers will be able to deploy Synopsys.ai Copilot on air-gapped on-prem environments, leveraging the accelerated computing performance of NVIDIA DGX systems.
The Synopys.ai Copilot capability is currently available for early-access trials.
With automotive design teams building and validating their products in the digital world before manufacturing them in the real world, Synopsys is now collaborating with NVIDIA to integrate Synopsys’ digital twin solutions with the NVIDIA Omniverse platform to reduce costs and time to market and improve the safety of software-defined and increasingly autonomous vehicles.
Synopsys systems software, virtual ECU and electronics digital twin fabric will be connected with Omniverse, a development platform for building interoperable 3D applications for industrial workloads.
Synopsys’ virtual prototyping solutions provide automotive engineering teams with a digital twin of the electronic system of the car to enable the development, testing and validation of the car’s software and electronic systems. Omniverse delivers physically based visualization and simulation of environmental factors.
Together, these capabilities can provide engineering teams with a digital twin of both the vehicle electronics and the environment, allowing them to test and validate embedded software, safety and autonomy features well ahead of production.
Synopsys said that it expects to begin engaging with lead customers on the solution in the second half of 2024, with general availability expected in 2025.
