NVIDIA Corporation's NVDA annual GPU Technology Conference is virtual this year and takes place May 19. Ahead of the conference, CEO Jensen Huang's keynote was made available on YouTube Thursday.
Nvidia's Largest Leap In Performance With A100 Data Center GPU
The Santa Clara, California-based company unveiled products based on its 7nm GPU architecture named Ampere. The company said the first Ampere-based data center GPU, A100, is in full production and shipping to customers worldwide.
The eighth generation of GPU is aimed at unifying AI training and inference and boosting performance by up to 20x over its predecessors, and it's also built for data analytics, computing and cloud graphics, according to the company. The previous version was the Volta architecture launched in 2017.
Microsoft Corporation MSFT is among the first to adopt Nvidia A<100 GPUs.
"Azure will enable training of dramatically bigger AI models using NVIDIA's new generation of A100 GPUs to push the state of the art on language, speech, vision and multi-modality," said Mikhail Parakhin, a Microsoft vice president.
Nvidia's Huang said that for the first time, "scale-up and scale-out workloads can be accelerated on one platform. NVIDIA A100 will simultaneously boost throughput and drive down the cost of data centers."
Nvidia's DGX A100 System
Nvidia also unveiled the DGX A100 AI system built on its A100 GPU, which the company said offers unprecedented compute density, performance and flexibility in the world's first five petaFLOPS AU system.
"NVIDIA DGX A100 features the world's most advanced accelerator, the NVIDIA A100 Tensor Core GPU, enabling enterprises to consolidate training, inference, and analytics into a unified, easy-to-deploy AI infrastructure that includes direct access to NVIDIA AI experts," according to the chipmaker.
See also: Why Nvidia Has A New Street-High Price Target
Nvidia Expands Edge Computing Capabilities With EGX A100
Nvidia also launched the EGX A100, which combines a A100 GPU with a Mellanox ConnectX-6 Dx SmartNIC network card. The company also announced the EGX Jetson Xavier NX for micro-edge servers.
"Servers powered by the EGX A100 can manage hundreds of cameras in airports, for example, while the EGX Jetson Xavier NX is built to manage a handful of cameras in convenience stores," Nvidia said.
EGX A100 is the first AI product based on the Ampere architecture, which Nvidia said can help an organization's servers carry out real-time processing and protection of the massive amounts of streaming data from edge sensors.
The Mellanox ConnectX-6 Dx SmartNIC can enable 25 Gbps speeds in data centers, according to Nvidia.
"The new SmartNIC includes two ports of 25 Gbps or a single port of 50 Gbps for Ethernet connectivity with PCIe Gen 3.0 and 4.0 host connectivity."
Nvidia's Other Reveals
Nvidia also unveiled its new Isaac Robotics Platform with automaker BAYERISCHE MOTO/S ADR BMWYY, which selected the platform for enhancing its automotive factories.
Nvidia also released the conversational AI service Nvidia Jarvis.
NVDA Price Action
Nvidia shares were trading down 0.32% at $310.20 at the time of publication Thursday. Related Link: Another Nvidia Analyst Boosts Price Target Ahead Of Q1 Report
Photo courtesy of Nvidia.
© 2024 Benzinga.com. Benzinga does not provide investment advice. All rights reserved.
Trade confidently with insights and alerts from analyst ratings, free reports and breaking news that affects the stocks you care about.
