Supermicro Pioneers NVIDIA BlueField-4 STX Storage Server for Enhanced AI Inference Performance
Supermicro, Inc. (NASDAQ: SMCI), a leading Total IT Solution Provider specializing in AI, Cloud, Storage, and 5G/Edge technologies, has made a significant announcement at NVIDIA GTC 2026. The company unveiled one of the industry's first Context Memory (CMX) storage servers, built on the newly introduced NVIDIA STX reference architecture. This innovative solution is specifically designed to accelerate the entire lifecycle of artificial intelligence applications, marking a major step forward in AI infrastructure.
Revolutionary STX Architecture and CMX Server Details
The BlueField-4 STX storage server integrates cutting-edge components, including the NVIDIA Vera CPU and NVIDIA ConnectX-9 SuperNIC. This server builds upon Supermicro's previous advancements, such as the Petascale JBOF all-flash array powered by NVIDIA BlueField-3. The CMX server is engineered to tackle complex AI workloads, particularly long-lived queries and multi-stage chain-of-thought agentic tasks. These require efficient access to prior and intermediate tokens associated with user queries, which are managed through a Key Value (KV) cache system orchestrated by NVIDIA Dynamo, NVIDIA's inference layer.
By leveraging the STX architecture, the CMX server not only accelerates result generation but also reduces power consumption. This is achieved by minimizing the need to recompute results when local storage for tokens is exceeded, offering a more sustainable and high-performance solution for AI-driven enterprises.
Leadership and Collaboration in AI Innovation
Charles Liang, president and CEO of Supermicro, emphasized the company's commitment to innovation, stating, "Supermicro continues to be first to market with new rack scale architectures designed to exceed the needs of a rapidly evolving AI Factory customer base. Building upon last year's introduction of the Petascale JBOF, where we proved the feasibility of a JBOF powered by NVIDIA BlueField-3 DPUs, we have developed the CMX storage server. Our prototype demonstrates our deep collaboration with NVIDIA and our dedication to delivering game-changing technologies."
As the STX solution enters the market, Supermicro will collaborate with software partners for porting and validation. The company's established relationships with top SSD providers like Micron, Samsung, and Phison will facilitate rigorous testing to meet the specific requirements of the STX architecture.
Expanding AI Data Platform Solutions
At GTC 2026, Supermicro also announced seven AI Data Platform solutions based on the RTX PRO 6000 Blackwell Server Edition GPU. These solutions involve partnerships with storage leaders such as Cloudian, DDN, Everpure (formerly Pure Storage), IBM, Nutanix, VAST Data, and WEKA. The AI Data Platform enables enterprises to efficiently process data for AI workloads, further solidifying Supermicro's role in the AI ecosystem.
The CMX server is being showcased at Supermicro booth #1113 and the NVIDIA exhibit during NVIDIA GTC 2026, held from March 16 to 19. This event highlights the ongoing advancements in AI technology and Supermicro's proactive approach to infrastructure development.
About Supermicro and Its Global Impact
Supermicro, headquartered in San Jose, California, is a global leader in Application-Optimized Total IT Solutions. The company focuses on delivering first-to-market innovations for Enterprise, Cloud, AI, and 5G Telco/Edge IT Infrastructure. With in-house design and manufacturing across the US, Asia, and the Netherlands, Supermicro optimizes for total cost of ownership and environmental sustainability through its Green Computing initiatives.
The award-winning Server Building Block Solutions® portfolio allows customers to tailor systems to their exact workloads, supporting a wide range of form factors, processors, memory, GPUs, storage, networking, and cooling solutions. Supermicro's trademarks, including Server Building Block Solutions and We Keep IT Green, underscore its commitment to innovation and efficiency in the tech industry.
