Table 2. Performance increase for each FullMonteCUDA optimization over FullMonteSW.
| Optimization | Incremental Speedup |
|---|---|
| Naive | 2x |
| CUDA vector datatypes and math operations | 2.5x |
| Materials constant cache | 1.6x |
| Thread local accumulation cache | 1.3x |
Official websites use .gov
A
.gov website belongs to an official
government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you've safely
connected to the .gov website. Share sensitive
information only on official, secure websites.