Overview
What this challenge is about.
Port a CFD stencil kernel to CUDA with shared-memory tiling on an A100 and validate it to earn your verifiable certificate.
The scenario
The subcontractor (3 active aerospace-prime contracts, 6-person sim-tech team) bills by the design iteration; 4.5 hours per iteration caps how many they can run per contract per week.
The Brief
What you'll do, and what you'll demonstrate.
Port a Navier-Stokes stencil kernel from CPU + OpenMP to CUDA on an A100, hit >=60 percent peak FP64 throughput, and validate per-cell numerical equivalence within 1e-8.
Earning criteria — what you'll demonstrate
- Port a stencil kernel to CUDA with shared-memory tiling
- Tune occupancy + register usage without spills
- Validate per-cell numerical equivalence rigorously
- Read Nsight Compute output and act on it
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Aligned coursework coming soon.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Gpu Programming
Apply gpu programming to solve real industry problems and demonstrate production-level capability.
- Cuda
Apply cuda to solve real industry problems and demonstrate production-level capability.
- Parallelism
Apply parallelism to solve real industry problems and demonstrate production-level capability.
- Performance Optimization
Apply performance optimization to solve real industry problems and demonstrate production-level capability.
- Cpp
Apply cpp to solve real industry problems and demonstrate production-level capability.
- Benchmarking
Apply benchmarking to solve real industry problems and demonstrate production-level capability.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles: