Microsoft Azure users can now leverage the latest NVIDIA accelerated computing technology to train, deploy and build generative AI applications. The Microsoft Azure ND H100 v5 VMs, powered by NVIDIA H100 Tensor Core GPUs and NVIDIA Quantum-2 InfiniBand networking, are now available to customers across the U.S. This new offering arrives at a time when developers and researchers are increasingly utilizing large language models (LLMs) and accelerated computing to discover novel consumer and business use cases.
The NVIDIA H100 GPU, a key component of this new offering, delivers supercomputing-class performance through a series of architectural innovations. These include fourth-generation Tensor Cores, a new Transformer Engine designed to accelerate LLMs, and the latest NVLink technology that allows GPUs to communicate with each other at a staggering speed of 900GB/sec.
Build generative AI applications
Further enhancing the performance of this new offering is the inclusion of NVIDIA Quantum-2 CX7 InfiniBand. With a cross-node bandwidth of 3,200 Gbps, this technology ensures seamless performance across the GPUs at a massive scale, matching the capabilities of the world’s top-performing supercomputers.
The ND H100 v5 VMs are particularly well-suited for training and running inference for increasingly complex LLMs and computer vision models. These neural networks power the most demanding and compute-intensive generative AI applications, including question answering, code generation, audio, video and image generation, speech recognition, and more.
Other articles you may be interested in the subject of generative AI :
In terms of performance, the ND H100 v5 VMs have demonstrated their potential to further optimize AI applications, achieving up to 2x speedup in LLMs like the BLOOM 175B model for inference compared to previous generation instances.
The integration of NVIDIA H100 Tensor Core GPUs on Azure provides enterprises with the performance, versatility, and scale needed to supercharge their AI training and inference workloads. This combination streamlines the development and deployment of production AI, with the NVIDIA AI Enterprise software suite integrated with Azure Machine Learning for MLOps. This powerful combination delivers record-setting AI performance in industry-standard MLPerf benchmarks, marking a significant milestone in the field of AI application development.
Source: NVIDIA
Filed Under: Technology News, Top News
Latest TechMehow
Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, TechMehow may earn an affiliate commission. Learn about our Disclosure Policy.