Home Tags NVIDIA



ASUS ESC4000A-E10 2U AMD EPYC GPU Server Review

The ASUS ESC4000A-E10 combines a single socket PCIe Gen4 AMD EPYC 7002/ 7003 with 4x double-width or 8x single-width GPUs in a 2U server

Lenovo ThinkSystem SD650-N V2 Liquid Cooled Intel Ice Lake

The Lenovo ThinkSystem SD650-N V2 is designed as a liquid cooled dual 3rd Gen Intel Xeon Scalable (Ice Lake) and quad NVIDIA A100 solution

Top500 November 2020 Our New Systems Analysis

In our Top500 November 2020 new systems analysis we see a sharp drop in the number of new systems added which makes trends stand out more
NVIDIA DGX Station A100 Cover

NVIDIA DGX Station A100 Brings 320GB HBM2e to Desktops

The NVIDIA DGX Station A100 has a 4x GPU NVIDIA A100 80GB configuration housed in a tower workstation for deskside AI development

NVIDIA A100 80GB Brings Double the Memory

The new NVIDIA A100 80GB model doubles memory offering 1.28TB in a 16x GPU system 640GB in an 8x GPU system or 320GB in a 4x GPU system.
NVIDIA Selene DGX SuperPOD Aisle

NVIDIA DGX A100 640GB Edition Launched

Alongside the NVIDIA A100 80GB, we now have the NVIDIA DGX A100 640GB Edition that takes eight of these GPUs and doubles memory capacity

Dell Precision 3640 Review Intel Xeon W and NVIDIA Quadro

The Dell Precision 3640 is a well-built workstation with Intel Xeon W-1200 CPUs and AMD/ NVIDIA GPUs among its configuration options
NVIDIA Tesla A100 Cover

AWS EC2 P4d Scales to 4000 NVIDIA A100 GPUs with UltraClusters

Amazon AWS EC2 P4d instances utilize Intel Xeon Scalable CPUs and NVIDIA A100 GPUs that can scale to over 4000 GPUs in AWS EC2 UltraClusters

NVIDIA Quadro RTX 4000 Review A Versatile AI and Professional GPU

In our NVIDIA Quadro RTX 4000 review, we see why some consider this GPU an awesome choice for compute and tasks such as AI inferencing
Gigabyte G242 Z10 Tesla T4 Array

MLPerf 0.7 Inference Released Others Join to Promote NVIDIA and Intel

MLPerf 0.7 Inference was released with new tests and the start of OEM participation to support NVIDIA. We also see some start to edge and IoT inclusion