« press release »
NVIDIA and Global Computer Makers Launch Industry-Standard Enterprise Server Platforms for AI
SANTA CLARA, Calif., April 12, 2021 (GLOBE NEWSWIRE) — NVIDIA today introduced a new class of NVIDIA-Certified Systems™, bringing AI within reach for organizations that run their applications on industry-standard enterprise data center infrastructure.
NVIDIA A10 & A30 Tensor Core GPUs, Source: NVIDIA
These include high-volume enterprise servers from top manufacturers, which were announced in January and are now certified to run the NVIDIA AI Enterprise software suite — which is exclusively certified for VMware vSphere 7, the world’s most widely used compute virtualization platform.
Further expanding the NVIDIA-Certified servers ecosystem is a new wave of systems featuring the NVIDIA A30 GPU for mainstream AI and data analytics and the NVIDIA A10 GPU for AI-enabled graphics, virtual workstations and mixed compute and graphics workloads, also announced today.
“AI is rapidly moving into mainstream use, accelerating demand for the infrastructure and software businesses require to deploy it at scale,” said Manuvir Das, head of Enterprise Computing at NVIDIA. “With NVIDIA AI Enterprise and VMware vSphere 7 on NVIDIA-Certified Systems, customers can now run virtualized AI applications on industry-standard servers — enabling hundreds of thousands of companies to host new AI services on their VMware platforms.”
Atos, Dell Technologies, GIGABYTE, H3C, Inspur, Lenovo, QCT and Supermicro are the first to offer NVIDIA-Certified mainstream servers supporting the NVIDIA EGX™ platform, enabling enterprises for the first time to run AI workloads on the same infrastructure used for traditional business applications.
Among the first incorporating these systems into their data centers are Lockheed Martin and Mass General Brigham.
NVIDIA and VMware’s collaboration provides customers an AI-ready enterprise platform to accelerate AI, container-based and traditional enterprise workloads, while also supporting virtualized AI applications with scale-out performance that is nearly indistinguishable from bare-metal servers.
“Customers don’t want AI silos – they want to run AI apps on their enterprise infrastructure for manageability, scalability, security and governance,” said Krish Prasad, senior vice president and general manager of the Cloud Platform Business Unit at VMware. “VMware and NVIDIA have teamed up so that customers can now evolve their existing enterprise infrastructure with an end-to-end AI-Ready Enterprise platform that’s easy to deploy and operate.”
NVIDIA-Certified EGX Systems Portfolio to Incorporate New Enterprise GPUs
Based on the NVIDIA Ampere architecture, the enterprise-class A30 delivers versatile performance at an optimal price for industry-standard servers. Each provides 24GB of HBM2 GPU memory and fast PCIe Gen 4 memory bandwidth while supporting four 6GB GPU instances with NVIDIA Multi-Instance GPU technology.
A30 supports a broad range of AI inference, training and traditional enterprise compute workloads. It can power AI use cases such as recommender systems, conversational AI and computer vision systems.
For AI training, its third-generation NVIDIA Tensor Cores support single-precision floating-point 32 calculations and an innovative new math mode known as TensorFloat-32, which boosts performance 20x over the previous-generation NVIDIA T4 GPUs.
The enterprise-grade NVIDIA A10 Tensor Core GPU powers accelerated graphics, rendering, AI and compute workloads in mainstream NVIDIA-Certified Systems. Built on the latest NVIDIA Ampere architecture, it provides 24GB of memory to accelerate the work of designers, engineers, artists and scientists.
Industry Leaders Embrace Virtualized AI
Industry innovators spanning healthcare, professional services, manufacturing and more are deploying NVIDIA-Certified Systems and NVIDIA and VMware’s AI-ready enterprise platform to power virtualized AI and data science.
“NVIDIA’s accelerated computing platform gives us the flexibility to support a broad range of mission-critical applications,” said Steven Walker, chief technology officer at Lockheed Martin. “From enabling real-time collaborative design and simulation, to deep learning capabilities that are revolutionizing predictive maintenance, cybersecurity and humanitarian assistance missions, NVIDIA-Certified Systems and software are critical to scaling infrastructure.”
“Virtualization is enabling healthcare systems to deliver services to clinicians and patients at scale, across radiology departments and facilities,” said Tom Schultz, director of Information Systems, Enterprise Medical Imaging, and Clinical Data Science at Mass General Brigham. “It has the potential to significantly increase the adoption of GPU-based AI applications. This allows for better utilization of technology infrastructure and minimizes the need for dedicated GPU systems for each project, which means AI can be applied more broadly to improve patient services.”
More than 20 NVIDIA-Certified Systems are available now from worldwide computer makers.
NVIDIA-Certified Systems featuring NVIDIA A30 and NVIDIA A10 GPUs will be available later this year from manufacturers.
NVIDIA AI Enterprise is available as a perpetual license at $3,595 per CPU socket. Enterprise Business Standard Support for NVIDIA AI Enterprise is $899 annually per license. Customers can apply for early access to NVIDIA AI Enterprise as they plan their upgrades to VMware vSphere 7 Update 2.
|NVIDIA Ampere Data Center Series|
|FP64 Tensor Core||–||10.3 teraFLOPS|
|FP32||31.2 teraFLOPS||10.3 teraFLOPS|
|TF32 Tensor Core||62.5 teraFLOPS | 125 teraFLOPS*||82 teraFLOPS | 165 teraFLOPS*|
|BFLOAT16 Tensor Core||125 teraFLOPS | 250 teraFLOPS*||165 teraFLOPS | 330 teraFLOPS*|
|FP16 Tensor Core||125 teraFLOPS | 250 teraFLOPS*||165 teraFLOPS | 330 teraFLOPS*|
|INT8 Tensor Core||250 TOPS | 500 TOPS*||330 TOPS | 661 TOPS*|
|INT4 Tensor Core||500 TOPS | 1,000 TOPS*||661 TOPS | 1321 TOPS*|
|RT Core||72 RT Cores||–|
2 decoder (+AV1 decode)
|1 optical flow accelerator (OFA)|
1 JPEG decoder (NVJPEG)
4 video decoders (NVDEC)
|GPU memory||24GB GDDR6||24GB HBM2|
|GPU memory bandwidth||600GB/s||933GB/s|
|Interconnect||PCIe Gen4 64GB/s||PCIe Gen4: 64GB/s|
Third-gen NVLINK: 200GB/s**
|Form factors||Single-slot, full-height, full-length (FHFL)||Dual-slot, full-height, full-length (FHFL)|
|Max thermal design power (TDP)||150W||165W|
|Multi-Instance GPU (MIG)||–||4 GPU instances @ 6GB each|
2 GPU instances @ 12GB each
1 GPU instance @ 24GB
|vGPU software support||NVIDIA Virtual PC, NVIDIA Virtual Applications, NVIDIA RTX Virtual|
Workstation, NVIDIA Virtual Compute Server
|NVIDIA AI Enterprise for VMware|
NVIDIA Virtual Compute Server
« end of the press release »