NVLink

Vultr Announces Addition of NVIDIA GH200 Grace Hopper Superchip to Its Cloud GPU Offerings for AI Training and Inference

Retrieved on: 
Monday, November 13, 2023
Software, Networks, Internet, Artificial Intelligence, Data Management, Technology, Semiconductor, Security, Cloud, GPU, LPDDR5X, HGX, Graph neural network, A40, PCI Express, Privacy, Acceleration, Genomics, NVIDIA GPU, NVLink, Computer graphics, Central processing unit, Computing, Graphics processing unit, CPU, NVIDIA, Satellite, H100, Drug discovery, Deep learning, HPC, TCO, Memory, Constant, ChatGPT, AI, A16, HBM3, Video game

Today, Vultr , the world’s largest privately-held cloud computing platform, announced the addition of the NVIDIA® GH200 Grace Hopper™ Superchip to its Cloud GPU offering to accelerate AI training and inference across Vultr’s 32 cloud data center locations .

Key Points: 
  • Today, Vultr , the world’s largest privately-held cloud computing platform, announced the addition of the NVIDIA® GH200 Grace Hopper™ Superchip to its Cloud GPU offering to accelerate AI training and inference across Vultr’s 32 cloud data center locations .
  • Following the launch of its first-of-its-kind GPU Stack and Container Registry , Vultr is providing cloud access to the NVIDIA GH200 Grace Hopper Superchip.
  • “The NVIDIA GH200 Grace Hopper Superchip delivers unrivaled performance and TCO for scaling out AI inference.
  • The NVIDIA GH200 Grace Hopper Superchip brings the new NVIDIA NVLink®-C2C to connect NVIDIA Grace™ CPUs with NVIDIA Hopper™ GPUs , delivering 7X higher aggregate memory bandwidth to the GPU compared to today’s fastest servers with PCIe Gen 5.

NVIDIA Announces DGX GH200 AI Supercomputer

Retrieved on: 
Monday, May 29, 2023
Cloud, NVLink, Meta, COMPUTEX, Fiat 500X, Helios, PCI Express, Microsoft, InfiniBand, Research, Language, Computing, Running, CPU, NVIDIA, Google Cloud, Architecture, AI, Library, Memory, GPU, DGX, Cryptocurrency, Video game, Medical device

TAIPEI, Taiwan, May 28, 2023 (GLOBE NEWSWIRE) -- COMPUTEX -- NVIDIA today announced a new class of large-memory AI supercomputer — an NVIDIA DGX ™ supercomputer powered by NVIDIA ® GH200 Grace Hopper Superchips and the NVIDIA NVLink® Switch System — created to enable the development of giant, next-generation models for generative AI language applications, recommender systems and data analytics workloads.

Key Points: 
  • TAIPEI, Taiwan, May 28, 2023 (GLOBE NEWSWIRE) -- COMPUTEX -- NVIDIA today announced a new class of large-memory AI supercomputer — an NVIDIA DGX ™ supercomputer powered by NVIDIA ® GH200 Grace Hopper Superchips and the NVIDIA NVLink® Switch System — created to enable the development of giant, next-generation models for generative AI language applications, recommender systems and data analytics workloads.
  • DGX GH200 is the first supercomputer to pair Grace Hopper Superchips with the NVIDIA NVLink Switch System, a new interconnect that enables all GPUs in a DGX GH200 system to work together as one.
  • DGX GH200 supercomputers include NVIDIA software to provide a turnkey, full-stack solution for the largest AI and data analytics workloads.
  • Also included is NVIDIA AI Enterprise , the software layer of the NVIDIA AI platform.

Supermicro Accelerates the Era of AI and the Metaverse with Top-of-the-Line Servers for AI Training, Deep Learning, HPC, and Generative AI, Featuring NVIDIA HGX and PCIe-Based H100 8-GPU Systems

Retrieved on: 
Tuesday, March 21, 2023
Cloud, Acceleration, Graphics processing unit, Deep learning super sampling, H100, DPU, Communication, Nvidia Omniverse, DL, 2U, 3D, DC, OVX, ML, PCI Express, InfiniBand, Central processing unit, AC, 5U, Running, OCP, HGX, Edge, NVIDIA, SAN, HPC, Cloud storage, L40, AV1, AI, TCO, NVLink, Total, GPU, Video game, L4, Supermicro

SAN JOSE, Calif., March 21, 2023 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for AI/ML, Cloud, Storage, and 5G/Edge, today has announced that it has begun shipping its top-of-the-line new GPU servers that feature the latest NVIDIA HGX H100 8-GPU system. Supermicro servers incorporate the new NVIDIA L4 Tensor Core GPU in a wide range of application-optimized servers from the edge to the data center.

Key Points: 
  • Supermicro servers incorporate the new NVIDIA L4 Tensor Core GPU in a wide range of application-optimized servers from the edge to the data center.
  • "With our new NVIDIA HGX H100 Delta-Next server, customers can expect 9x performance gains compared to the previous generation for AI training applications.
  • The Supermicro X13 SuperBlade® enclosure accommodates 20 NVIDIA H100 Tensor Core PCIe GPUs or 40 NVIDIA L40 GPUs in an 8U enclosure.
  • These new systems deliver the optimized acceleration ideal for running NVIDIA AI Enterprise, the software layer of the NVIDIA AI platform.

Supermicro Accelerates the Era of AI and the Metaverse with Top-of-the-Line Servers for AI Training, Deep Learning, HPC, and Generative AI, Featuring NVIDIA HGX and PCIe-Based H100 8-GPU Systems

Retrieved on: 
Tuesday, March 21, 2023
Cloud, Acceleration, Graphics processing unit, Deep learning super sampling, H100, DPU, Communication, Nvidia Omniverse, DL, 2U, 3D, DC, OVX, ML, PCI Express, InfiniBand, Central processing unit, AC, 5U, Running, OCP, HGX, Edge, NVIDIA, SAN, HPC, Cloud storage, L40, AV1, AI, TCO, NVLink, Total, GPU, Video game, L4, Supermicro

SAN JOSE, Calif., March 21, 2023 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for AI/ML, Cloud, Storage, and 5G/Edge, today has announced that it has begun shipping its top-of-the-line new GPU servers that feature the latest NVIDIA HGX H100 8-GPU system. Supermicro servers incorporate the new NVIDIA L4 Tensor Core GPU in a wide range of application-optimized servers from the edge to the data center.

Key Points: 
  • Supermicro servers incorporate the new NVIDIA L4 Tensor Core GPU in a wide range of application-optimized servers from the edge to the data center.
  • "With our new NVIDIA HGX H100 Delta-Next server, customers can expect 9x performance gains compared to the previous generation for AI training applications.
  • The Supermicro X13 SuperBlade® enclosure accommodates 20 NVIDIA H100 Tensor Core PCIe GPUs or 40 NVIDIA L40 GPUs in an 8U enclosure.
  • These new systems deliver the optimized acceleration ideal for running NVIDIA AI Enterprise, the software layer of the NVIDIA AI platform.

Cirrascale Cloud Services Offers Bare-Metal Cloud Servers and Clusters Powered by NVIDIA HGX H100 for Large-Scale AI and HPC Workloads

Retrieved on: 
Tuesday, March 21, 2023
Cloud, Cirrascale Cloud Services, NVLink, Deep learning super sampling, FP8, InfiniBand, Interest, HGX, NVIDIA, SAN, HPC, AI, Video game, H100, Hopper

SAN DIEGO, March 21, 2023 /PRNewswire-PRWeb/ -- Cirrascale Cloud Services®, the premier cloud services provider of the only AI Innovation Cloud containing the latest offerings from AI accelerator industry leaders, today announced the addition of NVIDIA® HGX H100 servers to its dedicated, multi-GPU deep learning cloud. The NVIDIA H100 Tensor Core GPU, powered by the Hopper architecture, delivers a massive leap in accelerated compute data center platforms, securely accelerating diverse workflows, from small enterprise workloads to exascale HPC and trillion-parameter AI.

Key Points: 
  • SAN DIEGO, March 21, 2023 /PRNewswire-PRWeb/ -- Cirrascale Cloud Services®, the premier cloud services provider of the only AI Innovation Cloud containing the latest offerings from AI accelerator industry leaders, today announced the addition of NVIDIA ® HGX H100 servers to its dedicated, multi-GPU deep learning cloud.
  • The NVIDIA HGX H100 server platform, as offered by Cirrascale, combines eight NVIDIA H100 GPUs with a high-speed interconnect powered by NVLink and NVSwitch technology to enable the creation of the world's most powerful scale-up servers.
  • Additionally, Cirrascale offers large-scale NVIDIA HGX H100 clusters built using NVIDIA Quantum-2 InfiniBand networking platform, so users can experience unmatched application performance across multiple servers.
  • NVIDIA HGX H100 servers are generally available on the Cirrascale Cloud Services platform.

Avicena Acquires microLED Fab Facility and Engineering Team From Nanosys

Retrieved on: 
Tuesday, October 18, 2022
Other Manufacturing, Mobile, Wireless, Technology, Semiconductor, Engineering, Aerospace, Software, Manufacturing, Networks, Hardware, CA, ASIC, Television, NVLink, Silicon, Xpu Há, PD, 2014 UCI Road World Championships, Multimedia, Investment, MicroLED, Partnership, HPC, Gallium nitride, CMOS, Acquisition, CEO, Bow, Interconnect, Artificial intelligence, Tablet, ICS, Video game console, Computer data storage, Avicenna, Nanosys

Avicena, the leader in high-performance microLED-based chip-to-chip interconnects, today announced that it has completed the acquisition of a microLED fabrication facility and associated engineering team from Nanosys.

Key Points: 
  • Avicena, the leader in high-performance microLED-based chip-to-chip interconnects, today announced that it has completed the acquisition of a microLED fabrication facility and associated engineering team from Nanosys.
  • Avicena had been using the Nanosys fab for the development of unique ultra-fast microLEDs.
  • The acquisition of the fab and associated engineering team significantly increases the development and manufacturing capabilities.
  • We thank the microLED fab team for their partnership in developing and manufacturing microLEDs as part of Nanosys and wish them well in their future with Avicena, said Jason Hartlove, President and CEO of Nanosys.

NVIDIA Unveils DRIVE Thor — Centralized Car Computer Unifying Cluster, Infotainment, Automated Driving, and Parking in a Single, Cost-Saving System

Retrieved on: 
Tuesday, September 20, 2022
AGX, Safety, Software, Deep learning super sampling, Engineering, Efficiency, CPU, Thor, Gao Conghui, NVLink, GTC, Compliance, Growth, SOC, Geely, Industry, MIG, NVDA, CEO, Technology, Architecture, Transport, Computer graphics, FP, NASDAQ, FP8, GPU, GLOBE, UNECE, IVI, ISO, DRIVE, Video game console, Cryptocurrency, NVIDIA

SANTA CLARA, Calif., Sept. 20, 2022 (GLOBE NEWSWIRE) -- GTC -- NVIDIA today introduced NVIDIA DRIVE™ Thor, its next-generation centralized computer for safe and secure autonomous vehicles.

Key Points: 
  • DRIVE Thor is the superhero of centralized compute, with lightning-fast performance to deliver continuously upgradable, safe and secure software-defined supercomputers on wheels.
  • DRIVE Thor replaces NVIDIA DRIVE Atlan and will be the follow-on to DRIVE Orin, which is currently in production and delivers 254 TOPS of performance.
  • Existing DRIVE Orin customers can take advantage of the platforms scalable architecture to easily transition current development efforts to DRIVE Thor.
  • Watch the NVIDIA GTC keynote address , in which Huang announced NVIDIA DRIVE Thor and other key automotive developments with IVI, mapping, simulation and more.

Avicena demonstrates ultra-fast microLED-array based interconnects for chip-to-chip communications at ECOC 2022

Retrieved on: 
Monday, September 19, 2022
Technology, Semiconductor, Engineering, Manufacturing, Software, Networks, Hardware, Other Science, Science, TBPS, MicroLED, Mountain View, California, Temperature, CMOS, ICS, 2013 UCI Road World Championships – Women's team time trial, ELS, NVLink, ECOC, PD, CEO, Technology, CA, Family, Application-specific integrated circuit, European Conference on Information Systems, Communes of Burundi, Gallium nitride, Conference, Interconnect, HPC, Multimedia, Bow, Xpu Há, Artificial intelligence, Computer data storage, Avicenna

Avicena, a privately held company in Mountain View, CA, is demonstrating its LightBundleTM multi-Tbps chip-to-chip interconnect technology at the European Conference for Optical Communications (ECOC) 2022 in Basel, Switzerland ( https://www.ecocexhibition.com/ ).

Key Points: 
  • Avicena, a privately held company in Mountain View, CA, is demonstrating its LightBundleTM multi-Tbps chip-to-chip interconnect technology at the European Conference for Optical Communications (ECOC) 2022 in Basel, Switzerland ( https://www.ecocexhibition.com/ ).
  • Highly variable workloads are driving the evolution of densely interconnected, heterogeneous, software-defined clusters of XPUs, Smart NICs, hardware accelerators, and high-performance shared memory.
  • Ever growing Artificial Intelligence (AI)/Machine Learning (ML) and High-Performance Computing (HPC) workloads are driving the need for interconnects with ultra-high bandwidth density, ultra-low power consumption, and low latency.
  • We have already demonstrated LightBundleTM links running at less than 1pJ/bit, says Bardia Pezeshki, founder and CEO of Avicena, Here at ECOC 2022 we are demonstrating individual microLED links running at 14Gbps.

Inspur AI Servers Achieve Record-Breaking Results in the Latest MLPerf v2.1 Inference Benchmarks

Retrieved on: 
Tuesday, September 13, 2022
Data Management, Other Technology, Technology, Artificial Intelligence, Software, NUMA, Software, Intel, BERT, Drugs and Cosmetics Rules, 1945, CPU, NVLink, Transformer, Deep learning, Data center, Technology, ImageNet, Solution, Cloud computing, Architecture, Data, Natural language processing, A100, Closed, System, Graphics processing unit, Research, AMD, GPU, AI, Medical device, Video game console, Risk management, Inspur

Inspur Systems, a leading data center, cloud computing and AI solutions provider, announced that Inspur AI servers achieved record-breaking results with massive performance gains in the newly-released MLPerf Inference v2.1 AI benchmark results.

Key Points: 
  • Inspur Systems, a leading data center, cloud computing and AI solutions provider, announced that Inspur AI servers achieved record-breaking results with massive performance gains in the newly-released MLPerf Inference v2.1 AI benchmark results.
  • 21 global companies and research institutions submitted more than 10,000 performance results for the Inference v2.1 benchmarks.
  • It is also marked the 4th time that an Inspur AI Server was the benchmark leader for the MLPerf inference BERT task.
  • Inspur is also the first to use the self-developed convolution merging algorithm plugin operator solution in the MLPerf Inference benchmarks.

Dihuni Announces NVIDIA GPU Parallelization and Optimization Services for Servers and Embedded Edge AI, Deep Learning, Machine Learning and HPC Performance

Retrieved on: 
Tuesday, August 30, 2022
Machine learning, Software, Digital transformation, CUDA, Digital, Deep learning, Data center, NVIDIA, HPC, Artificial intelligence, Internet, Solution, CPU, Data, NVLink, PyTorch, IOT, Multimedia, TensorFlow, Graphics processing unit, Algorithm, DL, Apache MXNet, GPU, AI, NVIDIA GPU, Video game console

MCLEAN, Va., Aug 30, 2022 /PRNewswire/ -- Dihuni, a leading Artificial Intelligence (AI), Data Center and Internet of Things (IoT) solutions company, today announced advanced engineering services to help customers extract maximum performance from NVIDIA® GPU-based AI, deep learning, machine learning and HPC infrastructure.

Key Points: 
  • "Dihuni has led the GPU server, workstation, and software space and now we are excited to extend our offerings to include NVIDIA GPU performance optimization services," said Pranay Prakash, Chief Executive Officer at Dihuni.
  • For a limited time, Dihuni will offer one-hour GPU performance optimization consulting at no cost to all of its Dihuni CognitX OptiReady GPU server customers.
  • Dihuni's GPU performance optimization services are available with NVIDIA H100 Tensor Core GPU (coming soon), NVIDIA A100 Tensor Core GPU, NVIDIA A40 GPU, NVIDIA A30 Tensor Core GPU, NVIDIA A10 Tensor Core GPU, NVIDIA A16 Tensor Core GPU and NVIDIA RTX A6000, NVIDIA RTX A5000 andNVIDIARTXA4000 GPUs.
  • To learn more about Dihuni's GPU performance optimization services and to sign up for one-hour free consulting, please visit https://www.dihuni.com/artificial-intelligence-ai-high-performance-compu... .