DDN at GTC Says Storage Appliance Doubles NVIDIA DGX Performance

Print Friendly, PDF & Email

CHATSWORTH, Calif. – March 22, 2022 – AI and multi-cloud data management company DDN today announced flash and hybrid data platforms for NVIDIA DGX POD and  DGX SuperPOD AI, analytics and deep learning computing infrastructure.

Powering thousands of NVIDIA DGX systems, including NVIDIA’s Selene and Cambridge-1 DGX SuperPOD systems, DDN offers AI data storage solutions for applications such as autonomous vehicles, natural language processing, financial modeling, drug discovery, academic research, and government security.

The DDN A3I AI400X2 system delivers real-world performance of more than 90 GB/s and 3 million IOPS to an NVIDIA DGX A100 system. Available with 250TB and 500TB all-NVMe usable capacity, and with the ability to scale orders of magnitude more, the DDN AI400X2 is the world’s most performing and efficient building block for AI infrastructures.

“DDN has been a market leader in AI, analytics and machine learning for many years and our collaboration with NVIDIA is leading the industry in performance, efficiency and ease of management at any scale,” said Dr. James Coomer, vice president of products, DDN. “With our next-generation flash and hybrid DDN AI400X2 storage systems, we are effectively doubling performance, improving ease of use and greatly expanding support for all AI users globally.”

NVIDIA DGX systems with DDN storage solutions have been implemented successfully by IT organizations worldwide. In 2021, DDN delivered more than 2.5 exabytes of AI, analytics and deep learning flash and hybrid storage solutions in the cloud and customers’ data centers. DDN expects to achieve significant growth in its AI business in 2022.

“NVIDIA DGX SuperPOD provides enterprises with a proven, turnkey AI infrastructure solution for powering their most transformative work,” said Charlie Boyle, vice president, DGX systems, NVIDIA. “From compute to networking to storage, every element of a DGX SuperPOD is selected to ensure it provides powerful performance, and DDN storage keeps pace with the needs of the most demanding AI workloads.”

DDN is working closely with NVIDIA on next-generation Reference Architecture documents that integrate DDN AI400X2 appliances with NVIDIA DGX A100 systems. Customers will be able to quickly deploy and scale AI turnkey systems using standard DGX POD and DGX SuperPOD configurations. Backed by NVIDIA’s leadership in accelerated computing and DDN’s leadership in AI data management at scale, these integrated systems will deliver the fastest path to AI implementation to customers across market segments and industries.