DDN’s reference architecture, integrated with NVIDIA BlueField-3 DPUs, utilizes the processing power of client-side DPUs to intelligently offload many critical functions such as S3 API calls, thereby delivering extraordinarily low latency at any scale. The added ability to distribute data operations to client or storage DPUs is critical for effective management and processing of AI data for generative AI clouds and data centers.
“Delivering AI acceleration is a full-stack problem and integrating our storage systems with BlueField-3 DPUs will allow us to deliver efficient storage services to a wider range of applications,” said Sven Oehme, chief technology officer at DDN. “This move toward disaggregating services will allow us to scale systems as generative AI and large language models continue to grow in model size and enterprise popularity at breakneck pace.”
DDN has a long history of working with NVIDIA, building solutions that deliver the highest value, performance and power efficiency for AI at rack- and data center-scale. As organizations deploy AI infrastructures in their data centers and in the cloud, the need for generative AI and AI frameworks to operate and scale reliably and cost-effectively has become increasingly important.
“DDN has seamlessly integrated NVIDIA BlueField-3 DPUs into its storage solutions, enabling organizations to scale their AI infrastructures with enhanced efficiency and delivering exceptional performance and value at rack- and data center-scale,” said Rob Davis, vice president of storage technology at NVIDIA.
