April 11, 2018
NVIDIA has announced availability of its new Quadro GV100 GPU (graphics processing unit). Based on the company’s Volta architecture, the GV100 is described as bringing designers, engineers and scientists “unprecedented capabilities in deep learning and simulation.”
NVIDIA says that the Quadro GV100 and its companion product, the Quadro vDWS for the data center, can be applied in applications in such diverse fields as automotive, architecture, engineering, entertainment and healthcare, where it can help users deploy deep learning-based research and development rapidly, accelerate AI (artificial intelligence) enhanced applications, enable photorealistic VR (virtual reality) and provide secure access.
The Quadro GV100 offers 7.4 TFLOPS (one trillion floating-point operations per second) double precision and 14.8 TFLOPS of single-precision performance paired with 118.5 TFLOPS of deep learning performance, according to NVIDIA. It is equipped with 32GB of high-bandwidth memory capacity, and two GV100 cards can be combined using NVIDIA NVLink interconnect technology to scale memory and performance.
Additional benefits include accelerated deep learning training and inferencing on a desktop workstation. This, NVIDIA explains, means that the GV100’s dedicated Tensor Cores and the ability to scale two GV100s for up to 64GB of HBM2 (high bandwidth memory, second generation) memory with NVIDIA NVLink provides the performance required for demanding deep learning training and inferencing applications. Further, the GV100 can provide “supercharged” rendering performance with accelerated denoising performance for ray tracing that provides fluid visual interactivity throughout the design process, according to the company.
NVIDIA says that among the other benefits of the GV100 is that it can help make the implementation of deep learning development initiatives easy through access to the NVIDIA GPU Cloud container registry. The registry, NVIDIA says, is a comprehensive catalog of GPU-optimized software tools for deep learning and high-performance computing on workstations.
The Quadro vDWS GPU for data centers offers new support for NVIDIA’s Tesla V100 GPUs. This, NVIDIA says, provides the power to address increasingly compute-intensive workflows and securely deliver workstation-class performance to any connected device. The company adds that with the Quadro vDWS, users can:
- run interactive, real-time simulations such as ANSYS Discovery Live;
- speed rendering times of photorealistic images up to 80% faster than with previous generations;
- leverage AI-enhanced applications for more fluid, visual interactivity throughout the design process; and
- work from anywhere, anytime, from any connected device, while data stays secure because it never leaves the data center.
For further details, click here.
Watch a video introduction on the Volta architecture.
Download a chart on choosing the right GPU for your applications.
Go here for more on the Quadro vDWS for the data center.
See why DE‘s Editors selected the NVIDIA Quadro GV100 GPU as their Pick of the Week.
Sources: Press materials received from the company and additional information gleaned from the company’s website.