Home Enterprise Explosive AI Adoption is Driving Energy-Efficient Development

Explosive AI Adoption is Driving Energy-Efficient Development

by Harold Fritts

As industries continue to adopt and integrate AI into the compute infrastructure, resources are taxed, and energy efficiency is becoming difficult to sustain. Overall, accelerated computing and the underlying technology deliver the efficiency to develop the next generation of innovation, with NVIDIA GPUs, the NVIDIA AI, and next-gen processors like the new 4th Gen Intel Xeon Scalable processors platform leading the way.

As industries continue to adopt and integrate AI into the compute infrastructure, resources are taxed, and energy efficiency is becoming difficult to sustain. Overall, accelerated computing and the underlying technology deliver the efficiency to develop the next generation of innovation, with NVIDIA GPUs, the NVIDIA AI, and next-gen processors like the new 4th Gen Intel Xeon Scalable processors platform leading the way.

With the launch of the 4th Gen Intel Xeon Scalable processor, NVIDIA and its partners are kicking off a new generation of energy-efficient AI-accelerated computing systems. When combined with NVIDIA H100 Tensor Core GPUs, these systems deliver dramatically higher performance, greater scale, and higher efficiency than prior generations.

New systems add performance and efficiency to Enterprise AI

New systems powered by NVIDIA and Intel will allow enterprises to run workloads 25x more efficiently than CPU-only data center servers. Delivering this performance per watt equates to using less power, which means more power for data centers to accelerate modern workloads.

The new generation of NVIDIA-accelerated servers speeds training and inferencing, boosting energy efficiency by 3.5x, which translates into actual savings, lowering TCO for AI data centers by 3x.

4th Gen Intel Xeon CPUs feature support for PCIe Gen 5

The new 4th Gen Intel Xeon CPU supports PCIe Gen 5, doubling the data transfer rates from CPU to NVIDIA GPUs and networking. The increased PCIe lanes provide a greater density of GPUs and high-speed networking within each server.

Faster memory bandwidth improves the performance of data-intensive workloads such as AI while networking speeds of up to 400 gigabits per second (Gbps) per connection support faster data transfers between servers and storage.

NVIDIA DGX H100 systems and servers from NVIDIA partners delivering NVIDIA DGX H100 systems and servers equipped with H100 PCIe GPUs come bundled with a license for NVIDIA AI Enterprise, an end-to-end, secure, cloud-native suite of AI development and deployment software, providing a complete platform for efficient enterprise AI.

NVIDIA DGX H100 Systems powered by NVIDIA Base Command

NVIDIA DGX H100 systems provide an optimized platform powered by the NVIDIA Base Command operating system designed for the accelerated data center. Each system features eight NVIDIA H100 GPUs, 10 NVIDIA ConnectX-7 network adapters, and dual 4th Gen Intel Xeon Scalable processors to deliver the performance necessary to build large generative AI models, large language models, recommender systems, and more.

Combined with NVIDIA networking, this architecture is expected to supercharge efficient computing at scale delivering up to 9x more performance than the previous generation and 20x to 40x more performance than unaccelerated X86 dual-socket servers for AI training and HPC workloads. With that kind of performance, what used to take a language model 40 days train could complete in less than two days.

The NVIDIA DGX H100 systems are the building blocks of the NVIDIA DGX SuperPOD, delivering up to one exaflop of AI performance, a leap in efficiency for large-scale enterprise AI deployment.

NVIDIA Partners Boost Data Center Efficiency 

For AI data center workloads, NVIDIA H100 GPUs enable enterprises to build and deploy applications more efficiently.

NVIDIA partners, including ASUS, Atos, Cisco, Dell Technologies, Fujitsu, GIGABYTE, Hewlett Packard Enterprise, Lenovo, QCT, and Supermicro, are expected to begin delivering the new portfolio of H100 GPUs and 4th Gen Intel Xeon Scalable CPUs servers soon.

Connecting servers with NVIDIA ConnectX-7 networking and Intel 4th Gen Xeon Scalable processors will increase efficiency and reduce infrastructure and power consumption. NVIDIA ConnectX-7 adapters support PCIe Gen 5 and connection speeds of 400 Gbps over Ethernet or InfiniBand, doubling networking throughput between servers and storage.

The adapters support advanced networking, storage, and security offloads. ConnectX-7 reduces the number of cables and switch ports needed, saving 17 percent or more on the power required for networking large GPU-accelerated HPC and AI clusters and contributing to the better energy efficiency of these new servers.

NVIDIA AI Enterprise Software delivers a full-stack AI solution

These next-generation systems are optimized for NVIDIA AI Enterprise Suite. Running on NVIDIA H100, NVIDIA AI Enterprise accelerates the data science pipeline. It streamlines the development and deployment of predictive AI models to automate essential processes and gain rapid insights from data.

With an extensive library of full-stack software, including AI workflows of reference applications, frameworks, pre-trained models, and infrastructure optimization, the software provides an ideal foundation for scaling enterprise AI success.

Engage with StorageReview

Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed