Supermicro Expands Integrated Open Compute Project Portfolio to Accelerate Rack-Scale AI Data Center Deployments
Integrated Open Compute Project (OCP) based portfolio spans rack-scale AI/HPC systems, ORv3 and ORW racks, 48V DC power, networking, and liquid cooling solutions to enable higher-density compute in existing and new data centers
Portfolio extends from infrastructure to intelligent operations with DCBBS Coolants, in-rack CDUs, switches, and Supermicro System Management Software Suite for infrastructure management
Supermicro DCBBS brings these OCP-based technologies together into integrated rack-scale solutions, simplifying and speeding deployment across compute, power, networking, and cooling for a faster Time-to-Online (TTO)
, /PRNewswire/ -- Super Micro Computer, Inc. (NASDAQ: SMCI ), an AI, Enterprise, Storage, and 5G/Edge Total IT Solution Provider featuring Data Center Building Block Solutions® (DCBBS) , today announced an expanded portfolio of Open Compute Project Accepted™ systems, racks, and rear door heat exchangers (RDHx). Supported by DCBBS, the solutions portfolio can be incorporated into complete end-to-end rack- and cluster-scale deployments.
Open Infrastructure for the Next Wave of AI
"The AI boom is redefining how data centers are designed, built, and deployed, and Supermicro is leading this transformation with DCBBS," said Charles Liang, president and CEO of Supermicro. "By integrating compute, networking, power, advanced cooling, software, and services into validated rack-scale solutions, we help customers simplify complexity and accelerate the path from innovation to production. DCBBS enables customers to reduce TTM and TTO by delivering essential AI infrastructure faster and more efficiently."
Supermicro's OCP solutions are now available, click here to learn more.
By leveraging OCP community best practices and DCBBS design and services, data center operators can increase compute density and energy efficiency for AI and HPC workloads based on an open and validated rack-scale foundation—from systems and busbar to rack-level and data center cooling—to bring high density solutions online more efficiently to accelerate Time-to-Market (TTM) and TTO.
"Customers need open, flexible infrastructure that can scale with the rapidly evolving demands of AI while fitting the realities of their data centers," said Dirk Van Slyke, vice president of Marketing, OCP. "By bringing OCP-accepted systems, racks, power and cooling together in validated rack-scale solutions, Supermicro is helping customers simplify integration, maximize facility capacity and efficiency, and accelerate the deployment of production AI infrastructure."
The ORv3 and ORW racks replace per-server power supplies and cables with a centralized 48V DC busbar, allowing customers to optimize their CPU and GPU systems for higher density compute racks. This density is further supported by Supermicro's DLC-2 Direct-to-Chip liquid cooling, which removes heat more efficiently from CPUs, GPUs, and other critical components, while residual heat can be removed from the rack by purpose-built ORv3 RDHx's without requiring additional retrofit of data hall air conditioning systems.
Supermicro RDHx units can be mounted to the rear of ORv3 and ORW racks and remove residual heat from server exhaust before it enters the data hall. The portfolio spans 10 kW to 120 kW of cooling capacity, giving operators a practical way to increase rack density in both new AI facilities and existing data centers. DC-powered models connect to the rack's busbar to reduce cabling, while designs include N+1 redundant fans to ensure uptime and a smart anti-condensation system to prevent moisture damage to other components inside the rack.
Operators can monitor and manage the entire portfolio of DCBBS components, through SuperCloud Composer ® using industry-standard protocols. Further, Supermicro's software, such as SuperCloud Director, SuperCloud Automation Assistant, and SuperCloud Developer Console, enables organizations to manage and operationalize all aspects of modern AI factories, enterprise data centers, and NeoCloud infrastructure, through a unified software control solution.
As part of DCBBS portfolio, Supermicro also delivers a comprehensive portfolio of AI networking solutions, ranging from air and liquid-cooled 800G and 1.6T Ethernet switches including next-generation Co-Packaged Optics (CPO) platforms built on Broadcom Tomahawk 6 switching silicon, optimized for EIA or ORv3 racks. The Supermicro Enterprise Advanced SONiC Networking Operating System gives customers the flexibility to select the networking architecture that best meets their performance, power, scalability, and deployment requirements as AI and large-scale data center environments continue to evolve.
Supermicro DCBBS will also include NVIDIA Spectrum-X Ethernet Photonics switch systems, the first co-packaged Ethernet switches with 200Gb/s SerDes. Built on the 102.4 Tb/s NVIDIA Spectrum-6 Ethernet ASIC , Spectrum-X Ethernet Photonics delivers 5x better power efficiency, 10x greater MTBI, 5x longer uptime for AI applications and 1.3x faster time to deployment, compared to pluggable transceiver-based networks. Spectrum-X Ethernet Photonics is in production now.
Portfolio Overview: Supermicro's broad solution portfolio incorporates many community-based standards, which enable faster and more optimized data centers to be brought online faster.
These include:
1 OU FlexTwin ™ Solution: A purpose-built rack-scale HPC solution featuring 1 OU 2-node systems powered by the latest Intel ® Xeon ® 6 or AMD EPYC ™ CPUs.
NVIDIA Vera Rubin NVL72: Supermicro's OCP contributed NVIDIA MGX 1.2 Rack Series is engineered exclusively for NVIDIA Vera Rubin NVL72 and NVIDIA Vera CPU platforms with Supermicro's DCBBS DLC-2 liquid-cooling stack support to streamline rack integration to accelerate deployment.