More NVIDIA GPU offerings on Google Cloud Platform

Google has expanded its NVIDIA GPU offerings on the Google Cloud Platform. These include:

  1. Performance boost with the public launch of NVIDIA P100 GPUs in beta
  2. NVIDIA Tesla K80 GPUs available on Google Compute Engine
  3. Introduction of sustained use discounts on both the Tesla K80 and P100 GPUs

According to a Google Cloud Platform blog, cloud GPUs can accelerate  workloads including machine learning training and inference, geophysical data processing, simulation, seismic analysis, molecular modeling, genomics and many more high performance compute use cases.

Continue reading “More NVIDIA GPU offerings on Google Cloud Platform”

NVIDIA and Microsoft to accelerate AI cloud computing

NVIDIA and Microsoft are working on a new hyperscale GPU accelerator that will provide hyperscale data centres with a fast, flexible path for artificial intelligence (AI).

The new HGX-1 hyperscale GPU accelerator is an open-source design released in conjunction with Microsoft’s Project Olympus.

HGX-1 does for cloud-based AI workloads what ATX — Advanced Technology eXtended — did for PC motherboards when it was introduced more than two decades ago. It establishes an industry standard that can be rapidly and efficiently embraced to help meet surging market demand.

Continue reading “NVIDIA and Microsoft to accelerate AI cloud computing”

RIKEN turns to NVIDIA supercomputer for deep learning research

RIKEN, Japan’s largest comprehensive research institution, will have a new supercomputer for deep learning research in April. Built by Fujitsu using 24 NVIDIA DGX-1 AI systems, the new machine will accelerate the application of artificial intelligence (AI) to solve complex challenges in healthcare, manufacturing and public safety.

Conventional high performance computing architectures are proving too costly and inefficient for meeting the needs of AI researchers. That’s why research institutions such as RIKEN are looking for GPU-based solutions that reduce cost and power consumption while increasing performance. Each DGX-1 combines the power of eight NVIDIA Tesla P100 GPUs with an integrated software stack optimised for deep learning frameworks, delivering the performance of 250 conventional x86 servers.

 

“We believe that the NVIDIA DGX-1-based system will accelerate real-world implementation of the latest AI technologies technologies as well as research into next-generation AI algorithms. Fujitsu is leveraging its extensive experience in high-performance computing development and AI research to support R&D that utilises this system, contributing to the creation of a future in which AI is used to find solutions to a variety of social issues,” said Arimichi Kunisawa, Head of the Technical Computing Solution Unit at Fujitsu.

 

 

Gunning for supercomputing supremacy in Japan

tsubame-3-0

Tokyo Institute of Technology plans to create Japan’s fastest AI supercomputer, which is will deliver more than twice the performance of its predecessor to slide into the world’s top 10 fastest systems.

Called Tsubame 3.0, it will use Pascal-based NVIDIA P100 GPUs that are nearly three times as efficient as their predecessors, to reach an expected 12.2 petaflops of double precision performance.

Tsubame 3.0 will excel in AI computation with more than 47 PFLOPS of AI horsepower. When operated with Tsubame 2.5, it is expected to deliver 64.3 PFLOPS, making it Japan’s highest performing AI supercomputer.

Continue reading “Gunning for supercomputing supremacy in Japan”

NVIDIA DGX SATURNV ranked most efficient supercomputer

nvidia_dgx_saturnvNVIDIA’s new DGX SATURNV supercomputer is ranked the world’s most efficient — and 28th fastest overall — on the latest Top500 list of supercomputers.

Powered by new Tesla P100 GPUs, it delivers 9.46 gigaflops/watt — a 42 percent improvement from the 6.67 gigaflops/watt delivered by the most efficient machine on the Top500 list released last June.

Compared with a supercomputer of similar performance, the Camphore 2 system, which is powered by Xeon Phi Knights Landing, SATURNV is 2.3x more energy efficient.hat efficiency is key to building machines capable of reaching exascale speeds — that’s 1 quintillion, or 1 billion billion, floating-point operations per second. Such a machine could help design efficient new combustion engines, model clean-burning fusion reactors, and achieve new breakthroughs in medical research.

Continue reading “NVIDIA DGX SATURNV ranked most efficient supercomputer”

CSIRO first in APAC to deploy NVIDIA DGX-1 supercomputers

david-kirk
NVIDIA Fellow David Kirk announcing the CSIRO deployment at GTCx Australia in Melbourne.

Australia’s federal research agency Commonwealth Scientific and Industrial Research Organisation (CSIRO) has become the first in Asia-Pacific to deploy the NVIDIA DGX-1 deep learning supercomputers.

Installed in CSIRO’s Canberra data centre, the two supercomputers will expand the capability of Australian scientists and broaden the science impact possibilities for the nation.

The NVIDIA DGX-1 is the world’s first deep learning supercomputer to meet the computing demands of artificial intelligence. It enables researchers and data scientists to easily harness the power of GPU-accelerated computing to create a new class of computers that learn, see and perceive the world as humans do.

Continue reading “CSIRO first in APAC to deploy NVIDIA DGX-1 supercomputers”

Pascal-based NVIDIA Tesla P100 delivers 12 times performance of previous generation accelerator

NVIDIA Tesla P100NVIDIA has introduced the NVIDIA Tesla P100 GPU, an advanced hyperscale data centre accelerator that can enable a new class of servers that can deliver the performance of hundreds of CPU server nodes.

Today’s data centres process large numbers of transactional workloads, such as web services. But they are inefficient at next-generation artificial intelligence and scientific applications, which require ultra-efficient, lightning-fast server nodes.

Based on the new NVIDIA Pascal GPU architecture, the Tesla P100 provides the performance and efficiency needed to power the computationally demanding applications. It delivers over a 12x increase in neural network training performance compared with a previous-generation NVIDIA Maxwell-based solution.

Continue reading “Pascal-based NVIDIA Tesla P100 delivers 12 times performance of previous generation accelerator”