Stephen Jones from Nvidia writes that Dynamic Parallelism in CUDA can speed up a wide range of tasks.
Dynamic Parallelism allows the GPU to operate more autonomously from the CPU by generating new work for itself at run-time, from inside a kernel. The concept is simple, but the impact is powerful: it can make GPU programming easier, particularly for algorithms traditionally considered difficult for GPUs such as divide-and-conquer problems.
Read the Full Story.
insideHPC in association with the technology analyst firm OrionX.net today announced the launch of the @HPCpodcast, featuring OrionX.net analyst Shahin Khan and Doug Black, insideHPC’s editor-in-chief. @HPCpodcast is intended to be a lively and informative forum examining key technology trends driving high performance computing and artificial intelligence. Each podcast will feature Khan and Blacks’ comments on the latest HPC news and also a deeper dive into a focused topic. In our first @HPCpodcast episode, we talk about a recent spate of good news for Intel before taking up one of the hottest areas of the advanced computing arena: new HPC-AI chips. You can find the @HPCpodcast on insideHPC and on Twitter. Here’s the RSS feed: http://orionx.net/category/audio-podcast/feed We welcome your suggestions [READ MORE…]
This whitepaper from One Stop Systems, describes how the requirements for AI in the field form a specific and distinct segment in the big, fast-growing edge computing market, separate from the familiar segments of edge data centers and the Internet of Things. One way to describe this emerging segment is “AI Transportables.”
Copyright © 2023
