General-purpose computing on graphics processing units (GPGPU) for modern computer systems
Modern computer systems rely heavily on parallel processing. Effectively, on the one hand almost any machine nowadays includes a multicore CPU, even mobile devices. On the other hand, parallel processing capacities are significantly augmented by the graphics processing unit (GPU): general-purpose computing on graphics processing units (GPGPU) is the typical example. In this research, relying on Microsoft’s DirectX 12 programming interfaces, we experiment a novel approach to enable parallel processing for graphical rendering on both the CPU and GPU for the popular Racket functional programming language and development environment (formerly PLT Scheme). And importantly as part of our objective, this is accomplished without compromising Racket’s ease of use and friendliness towards the programmer. In addition, the proposed framework has been empirically evaluated according to various use cases. Performance evaluations measured significant improvements with respect to execution time ([endif]--> speed-up in some cases), CPU utilisation time (reduced by as much as 80% in some scenarios), as well as the frame rate when using animated graphics.
Date and Time
Location
Hosts
Registration
- Date: 08 Dec 2017
- Time: 08:00 PM UTC to 09:00 PM UTC
-
Add Event to Calendar
- Reno, Nevada
- United States