NERP Tonight — NVIDIA’s CUDA: 192 Parallel Procesors

NERP is not exclusively Raspberry Pi, the small computer  interest group at Pumping Station:One in Chicago.  NERP meets  every other Monday at 7pm at Pumping Station:One, 3519 N. Elston  Ave. in Chicago.

Tonight at NERP, Sevin Straus will give an introduction to NVIDIA’s CUDA architecture. CUDA uses a cpu to farm out pieces of a task to the parallel processors in a video graphics chip (GPU). [more below]


[img: large-video-dynamic-parallelism-2-en.jpg]

Nvidia wants lots of developers to know about CUDA. To that end, they have put together a complete development environment. In the best of all worlds, the environment should be usable on Linux after simply running “”. It’s never really that simple. Sevin has put together a working develpment system targeting the Jeston development board. Tonight he’ll show us how he did it and some of the included demos.

A sense of what CUDA is about would include these thoughts collected from various parts of the CUDA website:

CUDA® is a parallel computing platform and programming model invented by NVIDIA. It enables dramatic increases in computing performance by harnessing the power of the graphics processing unit (GPU). A GPU consists of thousands of smaller, more efficient cores designed for handling multiple tasks simultaneously.

GPU-accelerated computing is the use of a graphics processing unit (GPU) together with a CPU to accelerate the compute-intensive portions of tan application to the GPU, while the remainder of the code still runs on the CPU. From a user’s perspective, applications simply run significantly faster.

– See more at:

Find NERP and Pumping Station:One
Doors open at 6:30pm. The next meeting is August 18th, 2014.
NERP is free and open to the public.
Ed Bennett ed @ kinetics and electronics com

Tags: electronics, embedded, NERP, Open Source, raspberry pi,
hackerspace, Beagle Bone, Element14, Pumping Station One


Share this!