Solidigm, a company specializing in storage technology for enterprise and hyperscale data environments, has launched a state-of-the-art facility designed for testing storage solutions in AI workloads. This new facility, known as the AI Central Lab, is located in Rancho Cordova, near Solidigm’s headquarters.
The AI Central Lab integrates powerful GPUs with high-density solid-state storage to analyze the real-world requirements of AI model training, inference, and evolving workload architectures. The lab aims to provide more extensive insights than typical component-level testing; it evaluates storage at the server, rack, and cluster levels under conditions that mimic large-scale data centers. By aligning with prevalent reference architectures, Solidigm allows customers and partners to test solutions using their solid-state drives in environments similar to actual production setups.
Avi Shetty, Senior Director of AI Ecosystems and Partnerships at Solidigm, noted that the lab meets an important industry demand. He explained that it combines top-tier GPUs with advanced storage infrastructure to facilitate comprehensive testing and innovation for customers and developers. This configuration opens capabilities that were once available only to a limited number of high-budget companies. He highlighted the importance of having storage in close proximity to GPU accelerators to maximize both utilization and performance.
The lab’s hardware setup reflects this philosophy, showcasing two groundbreaking test clusters. The first is recognized as the highest-performing storage test cluster, anchored by the Solidigm D7-PS1010 SSD, achieving a staggering per-node throughput of 116 GB/s during MLPerf Storage benchmark tests. The second cluster is noted for its density, with 192 Solidigm D5-P5336 SSDs, providing a combined storage capacity of 23.6 petabytes within just 16 rack units. Additional infrastructure includes NVIDIA B200 and H200 GPUs, 800 Gbps Ethernet networking, and storage servers from leading manufacturers.
The lab also focuses on power consumption and efficiency—a critical issue as AI infrastructure continues to grow. Engineers can benchmark system architectures against metrics like tokens per watt and tokens per dollar, increasingly crucial for assessing AI deployment economics. The facility is used to evaluate AI-specific workloads, including retrieval-augmented generation (RAG), vector database tuning, and key-value cache offloading, which require meticulous orchestration of storage and compute resources.
Solidigm is already utilizing the lab for collaboration with ecosystem partners. For example, their work with Metrum AI has demonstrated that moving certain AI data operations from memory to SSDs can cut DRAM usage by up to 57% during inference in RAG scenarios. Additionally, Solidigm is working with FarmGPU to enhance performance and efficiency strategies for AI infrastructure.
The AI Central Lab represents a significant move for Solidigm in showcasing how innovative storage design can elevate GPU productivity and ensure the efficiency of AI clusters. By merging high storage capacity with groundbreaking throughput and establishing a platform for testing real-world workloads, Solidigm is positioning storage as a crucial enabler of the next wave of AI systems.
Welcome to DediRock, your trusted partner in high-performance hosting solutions. At DediRock, we specialize in providing dedicated servers, VPS hosting, and cloud services tailored to meet the unique needs of businesses and individuals alike. Our mission is to deliver reliable, scalable, and secure hosting solutions that empower our clients to achieve their digital goals. With a commitment to exceptional customer support, cutting-edge technology, and robust infrastructure, DediRock stands out as a leader in the hosting industry. Join us and experience the difference that dedicated service and unwavering reliability can make for your online presence. Launch our website.