Performance Modeling Architect - Cache & NoC Innovations

NVIDIA Corporation; Santa Clara (CA); On-site; USD 184,000 - 288,000; Full time

Job summary

NVIDIA Corporation in Santa Clara is seeking a Performance Modeling Architect to define and improve CPU cache hierarchies and interconnects for automotive and data center systems.

You will develop cycle-accurate performance models (C++/SystemC), analyze bottlenecks, and evaluate coherency protocols. Collaboration with software and verification teams will optimize drivers and validate models against silicon.

Qualifications

• Master’s or Ph.D. in Computer Engineering, Electrical Engineering, or Computer Science (or equivalent) with 5+ years of experience.
• Strong understanding of CPU microarchitecture, memory consistency models, and cache coherency protocols.
• Proven experience in C++ or SystemC for cycle-accurate or functional modeling.
• Proficiency in Python or similar scripting languages for data processing and visualization.
• Understanding of NoC topologies (Mesh, Ring, Torus) and arbitration logic.

Responsibilities

• Develop and maintain high-fidelity, cycle-accurate performance models for coherent interconnects and large-scale caches.
• Model and analyze performance bottlenecks across automotive and data center scales.
• Evaluate performance impact of coherency protocols and snooping filters.
• Run industry benchmarks to drive architectural trade-offs and optimize drivers.

Skills

CPU microarchitecture

Memory coherency

C++/SystemC

Python scripting

NoC topologies

Education

Master's or PhD in relevant field

Tools

C++

SystemC

Job description

NVIDIA Corporation in Santa Clara is seeking a Performance Modeling Architect to define and improve CPU cache hierarchies and interconnects for automotive and data center systems.

You will develop cycle-accurate performance models (C++/SystemC), analyze bottlenecks, and evaluate coherency protocols. Collaboration with software and verification teams will optimize drivers and validate models against silicon.