HS Hyosung Infor Targets AI Service Operations Market With Arm Server and NPU Infrastructure
IT DAILY ·
✦ AI Summary
HS Hyosung Information Systems announced on the 9th that it will target the enterprise AI service operations market.
Based on a high-efficiency AI inference infrastructure that combines the low-power Arm server GreenCore with NPU, it is pursuing collaboration with Qualcomm, Furiosa AI, DeepX, Mobilint, and others.
NPU selection does not rely only on catalog performance, but analyzes customers' models, service types, number of users, response speed, power, and space conditions to present a suitable set of candidates.
HS Hyosung Information Systems announced its market push strategy on the 9th. Its target is the enterprise AI service operations market.
The company is entering the market based on a high-efficiency AI inference infrastructure that combines the low-power Arm server "GreenCore" with NPU.
To that end, it is pursuing broad collaboration with major domestic and overseas NPU companies, including Qualcomm, Furiosa AI, DeepX, and Mobilint. Based on this, it will provide infrastructure suited to customer environments.
When selecting an NPU, it does not rely solely on catalog performance as the only criterion. Instead, it analyzes the AI models and service types customers use, model size, the number of concurrent users, target response speed, data center power, and space conditions to present a suitable set of NPU candidates.
HS Hyosung Information Systems offers an option to maximize the advantages of NPU through its Arm server "GreenCore." GreenCore was jointly developed by HS Hyosung Information Systems and X-Access Lab, and X-Access Lab is a domestic Arm server manufacturer. GreenCore is a server lineup based on the CPU of the Arm architecture and offers low-power characteristics, low heat generation, and efficient many-core parallel processing.
GreenCore is intended for data centers facing limits in power, cooling, and rack space, as well as companies and institutions preparing on-premises AI services with security and data sovereignty in mind. HS Hyosung Information Systems directly installs the selected NPU in GreenCore.
It then integrates verification of hardware and software compatibility, AI model execution, and operational stability. If necessary, it works with AI solution companies to conduct a proof of concept (PoC) that reflects actual service environments. This reduces the burden on customers of having to directly compare and verify multiple NPU products and manufacturer-specific development tools.
HS Hyosung Information Systems supports the entire process of introducing AI inference infrastructure. The scope of support includes customer workload analysis, NPU selection, GreenCore server configuration, integrated verification, AI solution integration, PoC, deployment, and operations.
The company plans to continue strengthening its "HS Hyosung AI Platform" portfolio in the future. The platform comprises Arm servers, GPU, NPU, high-performance storage, data lakes, AIOps, and an AI solution partner ecosystem, and the company plans to organically combine these components. Through this, it aims to support customers' AI adoption, stable operations, and scaling.
Yang Jeong-gyu, CEO of HS Hyosung Information Systems, said that the center of AI competitiveness has shifted from model development to stable and efficient operation in real business environments. He added that, along with its plan to support a high-efficiency inference infrastructure combining GreenCore and various NPU, the company will help customers quickly implement AI services tailored to their workloads while considering constraints in cost, power, and space.
As a separate case, looking at HS Hyosung Information Systems' partnership history, the company signed an MOU with EDB in March, and the purpose of the MOU was enterprise data technology cooperation.
Source: IT DAILY · Kwon Young-seok
Original: https://www.itdaily.kr/news/articleView.html?idxno=241504
References
This article was produced with the help of an automated content generation algorithm.
Source: IT DAILY
View originalThis article was summarized and organized by BizCrush based on the original article from IT DAILY. For exact quotations and full details, please refer to the original article.