Liqid unveiled the Liqid UltraStack L40S reference architecture. This features Dell PowerEdge servers with 8 or 16 NVIDIA L40S GPUs that help customers meet emerging high GPU density demands. UltraStack can help accelerate next-generation workloads by delivering 35 percent higher performance, while ensuring cost savings with a 35 percent reduction in power consumption and a 75 percent reduction in software licensing costs when compared to lower GPU capacity servers.
In the rapidly evolving worlds of artificial intelligence (AI) and large language models (LLM), high performance computing (HPC) and graphics-intensive applications, the demand for high-density GPU systems has never been greater. However, access to high end GPUs, like NVIDIA’s H100 Tensor Core GPU, have higher lead times, and many customers need 8 or more GPUs in a single server.
“With the rising demand for GPU performance, Liqid steps up as a powerful ally,” said Sumit Puri, CEO & Cofounder, Liqid. “Our Liqid UltraStack reference architecture is a game-changer, built around Dell PowerEdge servers with NVIDIA L40S GPUs. It’s not just a concept – it’s a tangible, reliable solution readily available. This provides an unparalleled advantage for compute-intensive tasks, firmly placing you ahead of the curve.”
When implementing the Liqid UltraStack reference architecture with the Dell PowerEdge R760xa, customers can configure and deploy 16 NVIDIA L40S GPUs, offering more performance per server. This solution combines NVIDIA L40S GPUs, NVIDIA ConnectX-7 NICs, NVIDIA BlueField DPUs, and Liqid NVMe IO Accelerator cards into a single Dell PowerEdge R760xa server, redefining the limits of high-density GPU systems.
