NAMU AX and Cocolink Sign MOU on AI Server and Operations Platform Integration

NAMU AX announced on the 14th that it had signed a memorandum of understanding (MOU) with Cocolink to jointly promote domestic AI infrastructure. The signing ceremony was held at NAMU AX's headquarters, and the photo shows Cocolink CEO Lee Dong-hak and NAMU AX co-CEO Kim Hwa-jung.

The agreement aims to combine Cocolink's high-density AI servers with NAMU AX's GPU resource and AI model operation technologies to supply a domestically produced AI server appliance that integrates hardware and software.

The two companies agreed to cooperate on jointly planning and supplying domestic AI server appliances. They also plan to work together on technology integration in the agentic AI platform sector and on responding jointly to domestic AI data center construction projects.

Specifically, they plan to preinstall NAMU AX's AI infrastructure operations platform on Cocolink's 'Klimax' AI servers. They will also verify interoperability between 'Klimax' and NAMU AX's platform.

The infrastructure provided by Cocolink is a server based on Non-blocking PCIe Fabric. This technology enables high-density interconnection of multiple AI accelerators, including GPU and NPU.

NAMU AX applies functions for GPU resource allocation, isolation and scheduling, usage status management, AI model deployment, endpoint creation, and operational status monitoring. Through this, it supports shared GPU resource use by multiple users and models, and enables a single platform to handle everything from infrastructure management to model deployment and operation.

The first result of the agreement is a domestic AI server appliance. The first unveiling is scheduled for the 26th, at the '2026 2nd KOITA-KOSA Solution Day,' co-hosted by the Korea Industrial Technology Association (KOITA) and the Korea Artificial Intelligence and Software Industry Association (KOSA).

The two sides will also pursue integration with the enterprise agentic AI platform 'NAA (NAMU Agentic AI).' This is intended to provide AI services specialized for internal corporate data and operations, and will be developed into a domestic AI data center model that covers everything from infrastructure operations to the use of AI services.

The scope of application for this project is broad, ranging from manufacturing, including factory automation and visual inspection, to public sector, finance, healthcare, defense, and universities and research institutions that share GPU resources. Going forward, they plan to expand support to next-generation AI accelerators and models, and diversify product configurations by customer size and use case.

Cocolink CEO Lee Dong-hak said the MOU would allow Cocolink's hardware capabilities and NAMU AX's software technologies to be combined into a practical alternative for domestic companies and institutions that have relied on expensive imported infrastructure. He also said the company would pursue data sovereignty and TCO innovation at the same time and lead the domestic AI infrastructure ecosystem.

NAMU AX CEO Kim Hwa-jung said the competitiveness of AI infrastructure depends not only on hardware performance but also on seamless integration with software and ease of operation. He also stressed that, together with Cocolink, the company would present a domestic AI infrastructure model that can be used across a range of industrial sites.

NAMU AX also announced its second-quarter earnings results and disclosed the figures, saying consolidated revenue came to KRW 35.4 billion, consolidated operating profit was KRW 300 million, and consolidated net profit was KRW 200 million.

Source: IT DAILY · Kwon Yeong-seok
Original: https://www.itdaily.kr/news/articleView.html?idxno=240999

References

• https://www.itdaily.kr/news/articleView.html?idxno=240999