Story image

NVIDIA sets records with their enterprise AI

13 Dec 18

Backed by Google, Intel, Baidu, NVIDIA and dozens of technology leaders, the new MLPerf benchmark suite measures a wide range of deep learning workloads. 

Aiming to serve as the industry’s first objective AI benchmark suite, it covers such areas as computer vision, language translation, personalised recommendations and reinforcement learning tasks.

NVIDIA achieved the best performance in the six MLPerf benchmark results it submitted for. 

These cover a variety of workloads and infrastructure scale – ranging from 16 GPUs on one node to up to 640 GPUs across 80 nodes.

The six categories include image classification, object instance segmentation, object detection, non-recurrent translation, recurrent translation and recommendation systems. 

NVIDIA did not submit results for the seventh category for reinforcement learning, which does not yet take advantage of GPU acceleration.

A benchmark on which NVIDIA technology performed particularly well was language translation, training the Transformer neural network in just 6.2 minutes. 

NVIDIA engineers achieved their results on NVIDIA DGX systems, including NVIDIA DGX-2, featuring 16 fully connected V100 Tensor Core GPUs.

Performance on complex and diverse computing workloads takes more than chipsets. NVIDIA’s stack includes NVIDIA Tensor Cores, NVLink, NVSwitch, DGX systems, CUDA, cuDNN, NCCL, optimised deep learning framework containers and NVIDIA software development kits.

The software used to achieve NVIDIA’s MLPerf performance are available in the latest NGC deep learning containers.

The containers include the complete software stack and the top AI frameworks, optimised by NVIDIA. 

How enterprises use the containers: 

  • For data scientists on desktops, the containers enable research with NVIDIA TITAN RTX GPUs.
  • For workgroups, the same containers run on NVIDIA DGX Station.
  • For enterprises, the containers accelerate the application of AI to their data in the cloud with NVIDIA GPU-accelerated instances from Alibaba Cloud, AWS, Baidu Cloud, Google Cloud Platform, IBM Cloud, Microsoft Azure, Oracle Cloud Infrastructure and Tencent Cloud.
  • For organisations building on-premise AI infrastructure, NVIDIA DGX systems and NGC-Ready systems from Atos, Cisco, Cray, Dell EMC, HP, HPE, Inspur, Lenovo, Sugon and Supermicro supposedly put AI to work.
Fujitsu takes conservation prize for immersion cooling system
The prize was awarded for the Fujitsu Server PRIMERGY Immersion Cooling System that can reduce power consumption by up to 40%.
Why it’s critical to deploy emerging technologies in ways that will benefit society
“Technologies like 5G, quantum computing, artificial intelligence and biometrics are individually disruptive, but collectively they will completely rewrite global trust models."
Why 37% of enterprises believe in artificial intelligence
The number of enterprises implementing artificial intelligence grew 270% in the past four years and tripled in the past year, according to Gartner.
NetFoundry signs new partner to implement cloud solutions
NetFoundry AppWANs can supposedly facilitate enterprise application deployment and operations, such as SAP on Azure.
Common missteps on the path to enterprise mobility
Many organisations don’t think broadly enough when it comes to mobilising staff and workflows, focusing instead on only the most obvious users and needs.
Expert comment: Google fined US$57mil for GDPR breaches
The committee examining the breaches found two types of breaches of the GDPR.
McAfee customer choice for Cloud Access Security
“This is the second time that McAfee has been named... and we believe this demonstrates our ability to stay ahead of the pack.”
What it takes to be a successful leader during the ‘fourth industrial revolution’
"Leaders are more realistic about what it will take to succeed, and they appear particularly focused on societal impact and workforce development as two critical components of their future success."