WEKA launches AI storage cluster built on NVIDIA Grace CPU Superchips combining compute-accelerated storage.
WEKA, the AI-native data platform company, has unveiled the industry's first high-performance storage solution for the NVIDIA Grace CPU Superchip, previewed at Supercomputing 2024. The solution runs on a powerful new storage server from Supermicro powered by WEKA Data Platform software and Arm Neoverse V2 cores, utilizing the NVIDIA Grace CPU Superchip alongside NVIDIA ConnectX-7 and NVIDIA BlueField-3 networking. This collaboration between WEKA, NVIDIA, Arm, and Supermicro aims to deliver exceptional performance density and energy savings for enterprise AI deployments.
Today's AI and high-performance computing workloads demand lightning-fast data access, yet most data centers face increasing space and power constraints. The NVIDIA Grace CPU Superchip integrates the performance of a flagship x86-64 two-socket workstation into a single module, powered by 144 high-performance Arm Neoverse V2 cores that deliver 2x the energy efficiency of traditional x86 servers. The NVIDIA ConnectX-7 NICs and BlueField-3 SuperNICs feature purpose-built RDMA/RoCE acceleration, delivering high-throughput, low-latency network connectivity at up to 400Gb/s speeds. Connected by a high-performance custom-designed NVIDIA Scalable Coherency Fabric, this architecture delivers the performance of a dual-socket x86 CPU server at half the power.
The combination of WEKA's revolutionary zero-copy software architecture with the Supermicro Petascale storage server minimizes I/O bottlenecks and reduces AI pipeline latency, significantly enhancing GPU utilization and accelerating AI model training and inference. WEKA's AI-native architecture accelerates time to first token by up to 10x, while the Grace CPU's LPDDR5X memory architecture ensures up to 1 TB/s of memory bandwidth and seamless data flow. The WEKA Data Platform delivers 10-50x increased GPU stack efficiency and can shrink data infrastructure footprints by 4-7x while reducing carbon output—avoiding up to 260 tons of CO2e per PB stored annually and lowering energy costs by 10x.
"AI is transforming how enterprises around the world innovate, create, and operate, but the sharp increase in its adoption has drastically increased data center energy consumption, which is expected to double by 2026, according to the International Atomic Agency," said Nilesh Patel, chief product officer at WEKA. "WEKA is excited to partner with NVIDIA, Arm, and Supermicro to develop high-performance, energy-efficient solutions for next-generation data centers that drive enterprise AI and high-performance workloads while accelerating the processing of large amounts of data and reducing time to actionable insights."