Nvidia shows off the first A100 GPU systems built to handle AI workloads

As AI workloads can be complex and demanding, Nvidia has unveiled a new program that makes it easier for businesses to choose servers geared towards machine learning and analytics.

This new class of accelerated servers from the company's partners are given the Nvidia-Certified Systems branding as they are designed to power organization's journeys into AI and data analytics. In fact, the first wave of Nvidia-Certified Systems are now available to purchase and will offer a major performance boost in data centers around the world.

All of these systems are capable of handling AI workloads thanks to the inclusion of Nvidia A100 Tensor Core GPUs running on Nvidia Mellanox networks.

Nvidia Certified

(Image credit: Nvidia)

Nvidia-Certified Systems

So far, Dell Technologies, Gigabyte, Hewlett Packard Enterprise, Inspur and Supermicro are all shipping certified servers as Nvidia has collaborated with top OEMs from around the world to drive AI forward cross every industry.

The first systems using Nvidia A100 Tensor Core GPUs include the Dell EMC PowerEdge R7525 and R740 rack servers, GIGABYTE R281-G30, R282-Z96, G242-Z11, G482-Z54, G492-Z51 systems, HPE Apollo 6500 Gen10 System and HPE ProLiant DL380 Gen10 Server, Inspur NF5488A5 and the Supermicro A+ Server AS -4124GS-TNR and AS -2124GQ-NART.

In order to receive the Nvidia-Certified Systems branding, each server must pass tests involving deep learning training and inference, machine learning algorithms, intelligent video analytics and network and storage offload. These tests focus on real-world use cases and use popular AI frameworks and containers which are all available in the NGC catalog.

The Nvidia-Certified Systems badge gives customers confidence they are buying systems capable of meeting Nvidia's best design practices and additional partners are expected to release their own accelerated servers later this year.



from TechRadar - All the latest technology news https://ift.tt/39l1SOJ