Low-Power Sensing and Autonomy With NVIDIA Jetson TK1

Originally published at: https://developer.nvidia.com/blog/low-power-sensing-autonomy-nvidia-jetson-tk1/

Figure 1: simple TK1 block diagram NVIDIA’s Tegra K1 (TK1) is the first ARM system-on-chip (SoC) with integrated CUDA. With 192 Kepler GPU cores and four ARM Cortex-A15 cores delivering a total of 327 GFLOPS of compute performance, TK1 has the capacity to process lots of data with CUDA while typically drawing less than 6W…

Great article Dustin. Great tips on setting up the hardware through the OS on the device.

"the video stream is depacketized by the CPU into a buffer shared with the GPU,"

could you please provide a link to an introduction to implementation?
ie how does one actually get this done?

Here's a link to an example from the CUDA C Programming Guide v6 on using unified managed memory: http://docs.nvidia.com/cuda...

Anyways in the case of networking sockets, one would use cudaMallocManaged() to allocate the buffer passed to recv()/read() before calling CUDA kernel. Simplified psuedocode:

void* my_image = NULL;
cudaMallocManaged(&my_image, size_of_image);

read(my_socket, my_image, size_of_image);


Hi Dustin,
Great article, i am using TK1, its a wonderful board.
I was trying to get VisionWorks kit, but i cant find its download link any where, can you tell where can i get it?

thanks for the nice article... I'm interested mainly in the second case study: "Tiled Tegra"; could you give some more information about the hardware you used to connect the boards? Switches and cables to do that are already available on the market or you had to implement custom hardware? Could you provide a kind of "shopping list" to reproduce that configuration?

Thanks in advance and Best Regards

Hi Dustin,

As one of guys have mentioned before, I also have interests in Tiled Tegra and try to find necessary pcie switch and mini-pcie to pcie adapters. However, I cannot sure what sorts of those components I have to purchase in order to implement the Tiled Tegra.

Is it possible for you to let us know what PCIE switch or adapters you used for your Tiled Tegra ?