NVIDIA Harnesses Vera CPU to Speed Up Design of Next-Generation CPUs and GPUs

Summary

NVIDIA's Vera CPU, with 88 specialized cores and a high-bandwidth memory feature, accelerates verification and simulation for next-generation CPUs and GPUs by boosting critical EDA workloads up to 1.5x.

Why it matters

Faster EDA cycles directly accelerate chip production, allowing engineers to verify designs more quickly and explore alternatives. By optimizing CPU-heavy workloads, Vera reduces costly subsequent iterations and supports increasingly complex silicon development.

Highlights

• Up to 1.5x performance increase in Cadence Jasper and Synopsys VCS workloads.
• Integration of the 88 specialized Olympus cores with LPDDR5X and a scalable, consistent fab.
• Strategic collaboration with industry leaders Cadence and Synopsys for application profiling.

Critical take

The reported gains are based on early testing of selected workloads; broader, real-world efficiency improvements across the full EDA workflow have yet to be demonstrated.