In this video from GTC Japan 2014, Marc Hamilton from Nvidia presents: The Evolution of HPC.
NVIDIA’s development of CUDA allowed GPUs to move beyond graphics processing tasks to general purpose scientific computing. In addition to large research supercomputer centers, the so-called GP-GPU was quickly adopted for numerically intensive commercial applications ranging from manufacturing to finance to oil & gas exploration. Today, hundreds of the top applications in these fields are accelerated by GPUs. The next stop for GP-GPU computing is the exciting new field of machine learning for big data analytics. Using Convolution Neural Networks running on GPUs, it is now for the first time practical to process the huge amounts of unstructured text, voice, image, and video data available on the web to perform classification, clustering, regression, and recommendation. While the early adopters of Convolution Neural Networks have been the largest web search and social networking companies, the technology is quickly moving into the commercial segment as companies work to better understand the Massive Amounts of customer data available to them.