NVLink Takes GPU Acceleration To The Next Level

May 4, 2016 Timothy Prickett Morgan One of the breakthrough moments in computing, which was compelled by necessity, was the advent of symmetric multiprocessor, or SMP, clustering to make two or more processors look and act, as far as the operating system and applications were concerned, as a single, more capacious processor. With NVLink clustering for GPUs and for lashing GPUs to CPUs, Nvidia is bringing something as transformative as SMP was for CPUs to GPU accelerators. The NVLink interconnect has been in development for years, and is one of the “five miracles” that Nvidia CEO and co-founder Jen-Hsun Huang said at the GPU Technology Conference a month ago had to come together to make the “Pascal” GP100 graphics chip possible. The other miracles include the Pascal architecture itself, which…


Link to Full Article: NVLink Takes GPU Acceleration To The Next Level

Pin It on Pinterest

Share This

Join Our Newsletter

Sign up to our mailing list to receive the latest news and updates about homeAI.info and the Informed.AI Network of AI related websites which includes Events.AI, Neurons.AI, Awards.AI, and Vocation.AI

You have Successfully Subscribed!