AEWIN Unveils Flagship DC-MHS AI Servers Featuring Arivor's Advanced 2P DLC Solutions at OCP 2026

SAN JOSE, Calif., Oct. 6, 2026 /PRNewswire/ -- AEWIN Technologies of Qisda Group will showcase its latest modular systems based on the DC-MHS framework at OCP 2026, highlighting next-generation servers and two-phase direct liquid cooling (2P DLC) solutions for rack-scale architectures. Aligned with Qisda Group's strategic focus on AI Infrastructure, AEWIN leverages the Group's integrated expertise in computing, networking, storage, and thermal management to deliver flexible infrastructure optimized for energy efficiency, effectively empowering the evolving demands of AI factories.

AEWIN utilizes the cutting-edge AMD EPYC™ 9006 series Server CPUs to offer a comprehensive, high-performance server portfolio driven by superior computing power and PCIe Gen6 connectivity for the AI Era. As higher computing power leads to critical challenges of thermal management, AEWIN leverages the cutting-edge technology from its subsidiary to integrate its rack-scale 2P DLC solutions to handle high-density infrastructure exceeding 200kW per rack with reduced PUE and long-lifetime sustainability.

To achieve ultra-high compute density, BG18-A10710 integrates 2P DLC solutions to support 600W TDP CPU and two single-width 600W GPUs in 1U form factor while 2U AI server, BG28-A10710, delivers four dual-width FHFL PCIe 6.0 x16 slots to support 600W GPUs. As the latest addition to the portfolio, the flagship BG48-A20813 is a 4U AI server delivering up to eight dual-width FHFL GPUs custom-engineered to power complex Agentic AI and next-generation model training.

For data-intensive workloads, BS28-A10710 delivers 16x E3.S drive bays, while BX28-A10710 supports up to 12 U.2 NVMe SSDs or 24 SATA drives for various kinds of storage configurations supporting AI data pipelines and intelligent data management. With OCP DC-SCM2-based system management powered by the ASPEED AST2700 BMC and redundant 80 PLUS Titanium power supplies, these platforms provide a reliable foundation for AI inference, HPC, cloud-native, and neocloud deployments.

Through the modular DC-MHS-based architecture and rack-scale approach, AEWIN demonstrates its capability to build performant, scalable, and reliable servers for AI infrastructure. By uniting AEWIN's server expertise with Arivor's advanced 2P DLC technology and the broader Qisda Group technology ecosystem, the solution addresses the growing demand for AI compute density, high-speed connectivity, storage scalability, and thermal efficiency to help enterprises accelerate the deployment of sustainable, high-performance AI infrastructure.

For more information, please visit www.aewin.com.

Visit AEWIN at OCP Global Summit 2026, October 12–15, at Booth #D9.