NVIDIA’s AI Enterprise software program proven at Supercomputing ‘23 connects accelerated computing to giant language mannequin use instances.
On the Supercomputing ‘23 convention in Denver on Nov. 13, NVIDIA introduced expanded availability of the NVIDIA GH200 Grace Hopper Superchip for high-performance computing and HGX H200 Programs and Cloud Cases for AI coaching.
Leap to:
NVIDIA HGX GH200 supercomputer enhances generative AI and high-performance computing workloads
The HGX GH200 supercomputing platform, which is constructed on the NVIDIA H200 Tensor Core GPU, will probably be obtainable by means of server producers and {hardware} suppliers which have partnered with NVIDIA. The HGX GH200 is anticipated to start out transport from cloud suppliers and producers in Q2 2024.
Amazon Internet Providers, Google Cloud, Microsoft Azure, CoreWeave, Lambda, Vultr and Oracle Cloud Infrastructure will supply H200-based situations in 2024.
NVIDIA HGX H200 options the next:
- NVIDIA H200 Tensor Core GPU for generative AI and high-performance computing workloads that require large quantities of reminiscence (141 GB of reminiscence at 4.8 terabytes per second).
- Doubling inference pace on Llama 2, a 70 billion-parameter LLM, in comparison with the NVIDIA H100.
- Interoperable with the NVIDIA GH200 Grace Hopper Superchip with HBM3e.
- Deployable in any sort of information heart, together with on servers with current companions ASRock Rack, ASUS, Dell Applied sciences, Eviden, GIGABYTE, Hewlett Packard Enterprise, Ingrasys, Lenovo, QCT, Supermicro, Wistron and Wiwynn.
- Can present inference and coaching for the biggest LLM fashions past 175 billion parameters.
- Over 32 petaflops of FP8 deep studying compute and 1.1TB of combination high-bandwidth reminiscence.
“To create intelligence with generative AI and HPC functions, huge quantities of information should be effectively processed at excessive pace utilizing giant, quick GPU reminiscence,” mentioned Ian Buck, vp of hyperscale and HPC at NVIDIA, in a press launch.
NVIDIA’s GH200 chip is suited to supercomputing and AI coaching
NVIDIA will now supply HPE Cray EX2500 supercomputers with the GH200 chip (Determine A) for enhanced supercomputing and AI coaching. HPE introduced a supercomputing resolution for generative AI made up in a part of NVIDIA’s HPE Cray EX2500 supercomputer configuration.
Determine A
![Multiple NVIDIA GH200 chips working together.](https://www.techrepublic.com/wp-content/uploads/2023/11/tr_20231115-nvidia-high-performance-ai-chips-figure_a-770x556.jpg)
The GH200 contains Arm-based NVIDIA Grace CPU and Hopper GPU architectures utilizing NVIDIA NVLink-C2C interconnect know-how. The GH200 will probably be packaged inside methods from Dell Applied sciences, Eviden, Hewlett Packard Enterprise, Lenovo, QCT and Supermicro, NVIDIA introduced at Supercomputing ’23.
SEE: NVIDIA introduced AI training-as-a-service in July (TechRepublic)
“Organizations are quickly adopting generative AI to speed up enterprise transformations and technological breakthroughs,” mentioned Justin Hotard, govt vp and common supervisor of HPC, AI and Labs at HPE, in a weblog put up. “Working with NVIDIA, we’re excited to ship a full supercomputing resolution for generative AI, powered by applied sciences like Grace Hopper, which is able to make it simple for purchasers to speed up large-scale AI mannequin coaching and tuning at new ranges of effectivity.”
What can the GH200 allow?
Tasks like HPE’s present that supercomputing has functions for generative AI coaching, which might be utilized in enterprise computing. The GH200 interoperates with the NVIDIA AI Enterprise suite of software program for workloads akin to speech, recommender methods and hyperscale inference. It might be used together with an enterprise’s knowledge to run giant language fashions educated on the enterprise’s knowledge.
NVIDIA makes new supercomputing analysis heart partnerships
NVIDIA introduced partnerships with supercomputing facilities around the globe. Germany’s Jülich Supercomputing Centre’s scientific supercomputer, JUPITER, will use GH200 superchips. JUPITER will probably be used to create AI basis fashions for local weather and climate analysis, materials science, drug discovery, industrial engineering and quantum computing for the scientific neighborhood. The Texas Superior Computing Heart’s Vista supercomputer and the College of Bristol’s upcoming Isambard-AI supercomputer may even use GH200 superchips.
Quite a lot of cloud suppliers supply GH200 entry
Cloud suppliers Lambda and Vultr supply NVIDIA GH200 in early entry now. Oracle Cloud Infrastructure and CoreWeave plan to supply NVIDIA GH200 situations sooner or later, beginning in Q1 2024 for CoreWeave; Oracle didn’t specify a date.