The GPU Computing Study Group at the Tokyo Institute of Technology co-hosted the event which saw David Kirk from NVIDIA keynote.
The conference included 50 breakout sessions, various papers, and research posters from engineers, professors and industry experts. The conference’s award for best paper was given to Guanghao Jin, from the Tokyo Institute of Technology. His paper is titled “Data Management and Loop Controlling to Surpass Memory Capacity of GPU in OpenACC Framework.”
In Kirk’s address, however, he talks of the programming platform CUDA which is specifically designed for GPUs. The creation of CUDA was fueled by market trends towards realistic graphics (including physics simulation), computational graphics, and visual computing.
Some applications of CUDA include:
- Tokyo Tech using GPUs to help identify genomic k-mers in DNA sequences
- Waseda University creating neural networks of GPUs to simulate robotic adaptive behavior
- Hokkaido University simulating the fracture and deformation of biological matter to train surgeons
Kirk notes that “CUDA is used in various areas in our lives, such as powering drones to spot and avoid collisions on their own, enabling medical devices to pinpoint lesions with better clarity to detect breast cancer earlier, and allowing researchers to run complex atomic simulations to help identify new ways to combat HIV.”
Clearly, these GPUs can do more than play the latest line of Call of Duty.
Source NVIDIA Blog.