Wedoany.com Report on Mar 18th, Supermicro, Inc. announced the industry's first Context Memory (CMX) storage servers as part of the NVIDIA STX reference architecture unveiled at NVIDIA GTC 2026. STX is a new modular reference architecture from NVIDIA designed to accelerate the complete AI lifecycle.

"Supermicro continues to lead in introducing new rack-scale architectures designed to meet the needs of our rapidly growing AI factory customer base," said Charles Liang, President and CEO of Supermicro. "Building upon the Petascale JBOF (Just a Bunch of Flash) introduced last year, we demonstrated the viability of a JBOF powered by NVIDIA BlueField-3 DPUs and developed the CMX storage server. The prototype of our latest storage architecture showcases the level of our collaboration with NVIDIA and our commitment to being first to market with disruptive technology."
For more information on the new Supermicro storage servers built on the NVIDIA STX reference architecture, please visit: www.supermicro.com/en/solutions/ai-storage.
Leveraging the STX architecture, the CMX servers are designed to address the challenges of long-context AI queries and multi-step Chain-of-Thought agent workloads, which require access to previous and intermediate tokens related to a user's query. This solution both accelerates results and reduces the power consumption needed to recompute results when the required local storage for tokens is exceeded. This token storage, known as a Key-Value (KV) cache, is managed by NVIDIA's inference orchestration layer, NVIDIA Dynamo.
As STX solutions become available, Supermicro will engage in porting and validation efforts with these software partners and others. Furthermore, Supermicro's longstanding relationships with leading SSD vendors such as Micron, Samsung, Phison, and others will support testing tailored to specific STX architecture requirements.
At GTC 2026, Supermicro also announced seven AI Data Platform solutions based on the RTX PRO 6000 Blackwell Server Edition GPU, with partners including NVIDIA and storage providers such as Cloudian, DDN, Everpure (formerly Pure Storage), IBM, Nutanix, VAST Data, and WEKA. The AI Data Platform enables enterprises to process their data for AI workloads. The CMX servers will be showcased at the Supermicro booth #1113 and the NVIDIA exhibit area during the NVIDIA GTC conference from March 16-19, 2026.









