Making the Transition from NVIDIA A100 to H100: How We Built a Custom Solution

Published:
9.16.24
Author:
Amet Tempor
,
Do Ut Magna
,
Dolore Sit
,
GPU

Systems powered by the latest update to the NVIDIA HGX AI supercomputing platform support the NVIDIA H100 Tensor Core GPU, enabling an order-of-magnitude leap for unprecedented performance, scalability, and security.

Cirrascale Cloud Services®, the premier cloud services provider of the only AI Innovation Cloud containing the latest offerings from AI accelerator industry leaders, today announced the addition of NVIDIA® HGX H100 servers to its dedicated, multi-GPU deep learning cloud. The NVIDIA H100 Tensor Core GPU, powered by the Hopper architecture, delivers a massive leap in accelerated compute data center platforms, securely accelerating diverse workflows, from small enterprise workloads to exascale HPC and trillion-parameter AI.

“High-performance, large-scale accelerated computing is essential in today’s AI landscape, and NVIDIA is once again providing a breakthrough innovation with the H100 GPU,” said Mike LaPan, vice president, Cirrascale Cloud Services. “Utilizing NVIDIA Hopper architecture to deliver industry-leading conversational AI, the NVIDIA H100 GPU can speed up large language models by 30X over the previous generation.”The NVIDIA HGX H100 server platform, as offered by Cirrascale, combines eight NVIDIA H100 GPUs with a high-speed interconnect powered by NVLink and NVSwitch technology to enable the creation of the world’s most powerful scale-up servers. Leveraging the power of multi-precision Tensor Cores in H100, an eight-way HGX H100 provides over 32 petaFLOPS of FP8 deep learning compute performance.

Additionally, Cirrascale offers large-scale NVIDIA HGX H100 clusters built using NVIDIA Quantum-2 InfiniBand networking platform, so users can experience unmatched application performance across multiple servers.“AI and HPC have the potential to transform organizations and society by solving some of the world’s toughest challenges,” said Dave Salvator, director of accelerated computing products, NVIDIA. “NVIDIA’s accelerated computing platform and collaboration with cloud innovators such as Cirrascale push the boundaries of AI and HPC exploration to achieve a new wave of breakthroughs that benefit us all.”

NVIDIA HGX H100 servers are generally available on the Cirrascale Cloud Services platform. Interested customers and partners can visit https://cirrascale.com/h100 or call (888) 942-3800 to sign up for service.

Share This Article

Ready To Get Started?

Ready to take advantage of our flat-rate monthly billing, no ingress/egress data fees, and fast multi-tiered storage?

Get Started