NVIDIA

NVIDIA utilizes the advantage of a graphic card (for parallel computing operations) and offers solutions for accelerating applications from the fields of artificial intelligence, machine learning and high-performance computing (HPC). Alongside the hardware, it also uses a range of software including kits and tools for developers and a code developed in house to accelerate AI and HPC applications. As such, NVIDIA GPUs are the motor driving the rapid development of artificial intelligence.

NVIDIA entered the data center sector with traditional vGPU hardware and software products for virtual desktops, applications and workstations. Concepts such as GRID or Quadro (virtual workstations or CAD workstations) play an important role here, as do the virtualization technologies on which they are based, which meet the workload demands of modern workplaces and increase the user density on each server.

The graphics cards are fitted in the servers of our OEM providers, which include Dell, HPE, Lenovo, Fujitsu, Cisco and Supermicro.

In addition, NVIDIA has developed a platform specifically for artificial intelligence developments: Known as DGX systems, they are adapted to meet requirements in the areas of deep learning and artificial intelligence. Alongside the A100 GPUs, the DGX systems particularly stand out thanks to the provided software stack (GPU cloud, containers with finished frameworks, matching drivers), uniform support for the entire stack and their NVLink technology (GPU interconnect).

Why SVA?

SVA accompanied the development in the respective areas, even at an early stage. Because we made significant investments in a dedicated demo lab and trained the corresponding systems engineers and data scientists, we were able to gather experience early on and realize PoCs together with customers.