Editor’s Pick: GPU Designed for Data Center Enterprise Visualization

AMD introduces the Radeon PRO V620 GPU, a data center GPU featuring AMDs next-generation architecture. The company says the V620 can deliver immersive and intensive 3D visual workloads at scale in the cloud. 

The V620 features GPU partitioning and multi-stream hardware accelerated encoders. The GPU is equipped with 32GB of GDDR6 memory, RDNA 2 architecture and AMD Infinity Cache, which the company says allows the GPU to deliver low-latency content to multiple simultaneous users.

RDNA 2 is the graphics architecture also used in the Radeon PRO W6000 series of workstation graphics cards. The company says generation 2 of this architecture delivers up to 79% more performance than RDNA 1, and up to 179% better performance than the previous AMD graphics architecture, GCN.

Infinity Cache is a new cache design that enables high bandwidth performance at low power and low latency. AMD says it uses a technique called Temporal Reuse to optimize the iterative reuse of graphics data. This also enables data to be accessed virtually and instantaneously. 

New in AMD RDNA 2 Compute Unit is a high-performance raytracing acceleration architecture known as Ray Accelerators. The company says this design is a specialized hardware solution that efficiently handles the complex intersection of ray calculations designed to accelerate ray tracing beyond what software can provide. 

AMD says key use cases for the V620 include Desktop as a Service, Workstation as a Service, Machine Training, and various Visual Compute applications. 

The Radeon PRO V620 GPU supports Windows Server 2019 64-bit Edition, Linux Red Hat, Linux CentOS 8.3, and Ubuntu 20.04. The board draws a total power of 300W peak. The bus is PCIe 4.0 x16. It is a full height, double slot card. 

The AMD Radeon PRO V620 is available from AMD dealers. 

Sources: Press materials received from the company and additional information gleaned from the company’s website. 

More AMD Coverage

Accelerating Electric Vehicle Development with Multidisciplinary Simulation and High-Performance Computing

In this new Making the Case guide, learn how a unified approach to design and multidisciplinary simulation from Dassault Systèmes, combined with high-performance computing powered by AMD EPYC™ processors, can accelerate EV design.