BROWSE
Detail
Conference Paper
AI 반도체 제품 개발 동향 및 기술 발전 방향 : 데이터센터용 성능 비교를 중심으로
Cited - time in 
Share 
Share
- Issue Date
- 2023-06
- Citation
- 대한전자공학회 학술 대회 (하계) 2023, pp.1-4
- Publisher
- 대한전자공학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- The Korean government recently suggested a project for developing world-class level AI semiconductor products for data centers. This paper investigated the development trends for both domestic and foreign AI semiconductor products for data centers and compared the products’ performance. In detail, we compared the performance of ten products, decoupling AI training and inference. The performance comparison included comparing (1) measures―operating performance and power consumption―in dual axis and (2) AI semiconductor product’s operating performance in relation to maximum power consumption according to the release date. In conclusion, we outlined domestic and foreign AI semiconductor products" development trends and suggested some implications and directions for future development; Korean AI semiconductor industry needs to be competitive in AI training, facing a rapidly growing global AI technology such as generative AI.
- KSP Keywords
- Data center, Developing world, Development trends, Dual Axis, Future Development, Industry needs, Operating performance, Performance comparison, Power Consumption, Release Date, Semiconductor industry(SI)