NVIDIA revealed numerous infrastructure, {hardware}, and sources for scientific analysis and enterprise on the Worldwide Convention for Excessive Efficiency Computing, Networking, Storage, and Evaluation, held Nov. 17 to Nov. 22 in Atlanta. Key amongst these bulletins was the upcoming basic availability of the H200 NVL AI accelerator.
The latest Hopper chip is coming in December
NVIDIA introduced at a media briefing on Nov. 14 that platforms constructed with the H200 NVL PCIe GPU might be obtainable in December 2024. Enterprise clients can consult with an Enterprise Reference Structure for the H200 NVL. Buying the brand new GPU at an enterprise scale will include a five-year subscription for the NVIDIA AI Enterprise service.
Dion Harris, NVIDIA’s director of accelerated computing, mentioned on the briefing that the H200 NVL is good for information facilities with decrease energy — underneath 20kW — and air-cooled accelerator rack designs.
“Firms can fine-tune LLMs inside just a few hours” with the upcoming GPU, Harris mentioned.
H200 NVL exhibits a 1.5x reminiscence improve and 1.2x bandwidth improve over NVIDIA H100 NVL, the corporate mentioned.
Dell Applied sciences, Hewlett Packard Enterprise, Lenovo, and Supermicro will assist the brand new PCIe GPU. It can additionally seem in platforms from Aivres, ASRock Rack, GIGABYTE, Inventec, MSI, Pegatron, QCT, Wistron, and Wiwynn.
SEE: Firms like Apple are working exhausting to create a workforce of chip makers.
Grace Blackwell chip rollout continuing
Harris additionally emphasised that companions and distributors have the NV GB200 NVL4 (Grace Blackwell) chip in hand.
“The rollout of Blackwell is continuing easily,” he mentioned.
Blackwell chips are bought out via the subsequent 12 months.
Unveiling the Subsequent Section of Actual-Time Omniverse Simulations
In manufacturing, NVIDIA launched the Omniverse Blueprint for Actual-Time CAE Digital Twins, now in early entry. This new reference pipeline exhibits how researchers or organizations can speed up simulations and real-time visualizations, together with real-time digital wind tunnel testing.
Constructed on NVIDIA NIM AI microservices, Omniverse Blueprint for Actual-Time CAE Digital Twins lets simulations that typically take weeks or months be carried out in actual time. This functionality might be on show at SC’24, the place Luminary Cloud will present how it may be leveraged in a fluid dynamics simulation.
“We constructed Omniverse in order that every little thing can have a digital twin,” Jensen Huang, founder and CEO of NVIDIA, mentioned in a press launch.
“By integrating NVIDIA Omniverse Blueprint with Ansys software program, we’re enabling our clients to deal with more and more advanced and detailed simulations extra shortly and precisely,” mentioned Ajei Gopal, president and CEO of Ansys, in the identical press launch.
CUDA-X library updates speed up scientific analysis
NVIDIA’s CUDA-X libraries assist speed up the real-time simulations. These libraries are additionally receiving updates focusing on scientific analysis, together with adjustments to CUDA-Q and the discharge of a brand new model of cuPyNumeric.
Dynamics simulation performance might be included in CUDA-Q, NVIDIA’s improvement platform for constructing quantum computer systems. The objective is to carry out quantum simulations in sensible instances — resembling an hour as an alternative of a 12 months. Google works with NVIDIA to construct representations of their qubits utilizing CUDA-Q, “bringing them nearer to the objective of attaining sensible, large-scale quantum computing,” Harris mentioned.
NVIDIA additionally introduced the most recent cuPyNumeric model, the accelerated scientific analysis computing library. Designed for scientific settings that usually use NumPy applications and run on a CPU-only node, cuPyNumeric lets these initiatives scale to 1000’s of GPUs with minimal code adjustments. It’s at the moment being utilized in choose analysis establishments.