NVIDIA has pulled yet another trick out of its always-filled hat of technology goodies with the launch of Volta, the world’s most powerful GPU computing architecture. At his keynote address at GTC in San Jose, NVIDIA CEO Jensen Huang dubbed it “the next level of computer projects”.
Volta is created to drive the next wave of advancement in artificial intelligence (AI) and high performance computing.
The first Volta-based processor is the NVIDIA Tesla V100 data centre GPU, which brings extraordinary speed and scalability for AI inferencing and training, as well as for accelerating HPC and graphics workloads.
As NVIDIA’s seventh-generation GPU architecture, Volta is built with 21 billion transistors and delivers the equivalent performance of 100 CPUs for deep learning.
It is five times faster than Pascal, the current-generation NVIDIA GPU architecture, in peak teraflops, and 15 times over the Maxwell architecture, launched two years ago. This performance surpasses by a factor of four times the improvements that Moore’s law would have predicted.
New DGX lineup
Together with Volta launch, NVIDIA has introduced a new lineup of NVIDIA DGX AI supercomputers with unmatched computing performance to advance the world’s most challenging AI research.
Jensen pointed out that the DGX -1 is an essential instrument of AI research. “What used to take a week now takes a shift,” he said.
The DGX-1 replaces 400 servers and offers 960 tensor TFLOPS. It will ship in Q3 and cost US$149,000.
Featuring NVIDIA Tesla V100 data centre GPUs based on the NVIDIA Volta architecture and a fully optimised AI software package, the systems deliver groundbreaking AI computing power three times faster than the prior DGX generation, providing the performance of up to 800 CPUs in a single system.
The NVIDIA Volta architecture-based DGX portfolio includes the NVIDIA DGX-1 AI supercomputer for data centre deployments and a new personal supercomputing workstation, the NVIDIA DGX Station.
Both systems benefit from the integrated NVIDIA GPU Cloud Deep Learning Stack delivered over the newly launched NVIDIA GPU Cloud.
The new NVIDIA DGX Station is the world’s first personal supercomputer for AI development, with the computing capacity of 400 CPUs, consuming nearly 40x less power, in a form factor that fits neatly deskside.
Engineered for peak performance and deskside comfort, the DGX Station is the world’s quietest workstation, drawing one-tenth the noise as other deep learning workstations. Data scientists can use it for compute-intensive AI exploration, including training deep neural networks, inferencing and advanced analytics.