#USA #San Jose #AI servers #AEWIN Technologies #OCP Summit

AEWIN to Reveal Advanced AI Servers at OCP 2026 Summit

During the upcoming Open Compute Project (OCP) Summit 2026, AEWIN Technologies, a subsidiary of Qisda Group, is set to unveil their latest advancements in AI server technology. This introduction will revolve around their flagship systems built on the modular data center hardware architecture (DC-MHS), which are complemented by the sophisticated two-phase direct liquid cooling (2P DLC) solutions developed by Arivor.

Taking place in San Jose, California, from October 12 to 15, the OCP Summit is a crucial gathering for leaders in the computing industry, focusing on innovative technologies that enhance data center efficiency and performance. AEWIN's participation highlights their commitment to AI infrastructure, aiming to meet the growing demands of AI engine workloads and high-density computing environments.

Pioneering Modular Systems for High-Performance AI

AEWIN's approach to modular systems represents a significant evolution in server technology. They are integrating AMD EPYC™ 9006 series processors, which enable top-tier performance and enhanced connectivity through PCIe Gen6. This configuration is specifically designed to cater to the needs of modern AI applications that require substantial computational power.

As the demand for intense computational capabilities increases, AEWIN addresses thermal management challenges by leveraging advanced 2P DLC technologies. These systems are capable of supporting high-density infrastructures that surpass 200 kW per rack, all while improving energy efficiency indicators and overall sustainability.

The showcase will include several server models, such as the BG18-A10710, which features a 1U format capable of housing a 600W processor and two 600W single-width GPUs, thereby achieving ultra-high computing density. Furthermore, the new BG48-A20813 model, presented in a 4U format, accommodates up to eight double-width GPUs, making it suitable for complex AI tasks and next-generation model training.

Versatile Configurations to Meet Diverse Needs

AEWIN recognizes that data-intensive workloads also require specific storage configurations. Their offerings include the BS28-A10710 model, equipped with 16 E3.S disk bays for substantial storage capabilities, and the BX28-A10710, which supports up to 12 NVMe SSDs or 24 SATA drives. This flexibility ensures that enterprises can tailor their infrastructure to suit the intricate data pipelines involved in AI operations.

Additionally, AEWIN's server platforms incorporate a management system based on OCP's DC-SCM2 standards, optimized through ASPEED AST2700 management processors and 80 PLUS Titanium redundant power supplies. This level of reliable design not only supports high-performance computing configurations but also enhances AI inference deployments and cloud-native applications.

Enabling Sustainable and High-Performance AI Infrastructure

AEWIN's modular DC-MHS architecture, combined with Arivor’s superior 2P DLC technology, underscores the company's capacity to develop scalable and efficient AI server solutions. These innovations provide businesses with the tools necessary to accelerate the deployment of sustainable, high-performance AI infrastructures, accommodating the growing computational demands and thermal efficiencies that contemporary enterprises require.

To learn more about AEWIN's groundbreaking technologies, visit www.aewin.com. Don't miss the opportunity to experience these advancements firsthand at booth D9 during the OCP 2026 Summit!