[ad_1]
Nvidia has sleepless its new Ampere-based high-powered GPUs for enterprise desktop and data center workloads.
The Nvidia RTX A6000 and Nvidia A40 GPUs are succeeding the company’s current Quadro RTX 8000 and 6000 cards.
“Built on the NVIDIA Ampere architecture, the GPUs feature new RT Cores, Tensor Cores, and CUDA that accelerate graphics, rendering, computing and AI significantly faster than previous generations,” said Nvidia.
GPUs are designed for use in high-demand applications outside of gaming or mainstream design and video playback.
Nvidia explained that among their use cases, the A6000 and A40 will allow engineers to develop innovative products, designers will create next-generation buildings, and scientists will discover advancements from anywhere in the world.
Nvidia didn’t reveal exact specs like clock speeds, but did claim the following hardware and feature improvements:
- Second generation RT cores – Offers up to twice the performance of the previous generation, plus simultaneous ray tracing, shading and computation.
- Third generation tensioner cores Provides up to 5 times the performance of the previous generation, up to 10 times in short supply, with support for the new TF32 and BF16 data formats.
- New CUDA colors – Delivers up to twice the performance of the previous generation FP32 for significant increases in graphics and computing.
- 48 GB GPU memory – Largest memory available in a single GPU, expandable to 96GB using NVLink to connect two GPUs.
- Virtualization – The addition of NVIDIA virtual GPU software, such as the NVIDIA Virtual Workstation, can support graphics workloads and powerful virtual workstation instances at scale for remote users, enabling larger workflows for workloads. high-end computing, artificial intelligence and design work.
- PCIe Gen 4 Provides twice the bandwidth of the previous generation, accelerating data transfers to the GPU for data-intensive workloads such as data science, hybrid rendering, and video streaming on PCIe Gen-enabled servers and workstations 4 such as Lenovo ThinkStation P620.
Early adopters
Some examples from several of the early Nvidia-named RTX A6000 GPU users include:
- Leading architecture firms Kohn Pedersen Fox Associates and HKS, Inc.
- Oscar-winning visual effects studios, Digital Domain and Framestore.
- Automobile manufacturer and electric mobility company Groupe Renault.
- Predator Cycling custom carbon bike manufacturer.
Nvidia noted that the RTX A6000 will be available from channel partners including PNY, Leadtek, Ingram Micro, Ryoyo, and on nvidia.com starting in mid-December.
The A6000-based workstations are expected to be released by major system manufacturers, including Boxx, Dell, HP, and Lenovo, in early 2021.
The A40-based servers will be available from Cisco, Dell, Fujitsu, Hewlett Packard Enterprise, and Lenovo around the same date.
The images below show the layouts of the Nvidia RTX A6000 and Nvidia A40 cards.
Nvidia RTX A6000
Nvidia A40
Now Read: Nvidia GeForce RTX 3070 Launch Delayed
[ad_2]
