At the IEEE High Performance Extreme Computing Conference (HPEC ’14) in September, Massimiliano Fatica and Everett Phillips from NVIDIA presented a paper Synthetic Aperture Radar imaging on a CUDA-enabled mobile platform. The imaging was done on a Jetson TK1! The abstract:
This paper presents the details of a Synthetic Aperture Radar (SAR) imaging on the smallest CUDA-capable platform available, the Jetson TK1. The results indicate that GPU accelerated embedded platforms have considerable potential for this type of workload and in conjunction with low power consumption, light weight and standard programming tools, could open new horizons in the embedded space.
The reason I think it’s worth a read is 1) of course there’s the whole radar thing, and synthetic aperture at that, but also 2) is the discussion of taking algorithms that are implemented on traditional CPU architectures and converting it to run in parallel on the GPU.
I don’t want to spoil it for you, but it goes from initially being way slow to way fast.
A fun read!