SK Hynix AI-Specific Computing Memory Solution: From AiM Device to Heterogeneous AiMX-xPU System for Comprehensive LLM Inference
•Recap Accelerator-in-Memory (AiM) & AiMX •System Extensions of AiMX Card for Datacenter •AiM & AiMX for On-device AI •Design Choices for Future AiM/AiMX •Conclusion
Authors
- Hyunha Joo
- Guhyun Kim (ORCID: https://orcid.org/0000-0001-7076-2818)
- Woojae Shin (ORCID: https://orcid.org/0000-0002-6155-3712)
- Yousub Jung
- Il Kon Kim (ORCID: https://orcid.org/0000-0002-2260-5241)
- Byeongju An
- Haerang Choi (ORCID: https://orcid.org/0000-0002-8933-6226)
- Yonghoon Park
- Euicheol Lim (ORCID: https://orcid.org/0000-0002-8910-533X)
- B. S. Yang (ORCID: https://orcid.org/0000-0001-5877-6096)
- Jeongbin Kim (ORCID: https://orcid.org/0000-0002-5950-0056)
- Nahsung Kim
- Jae‐Han Park (ORCID: https://orcid.org/0000-0001-5837-3131)
- Yosub Song
- Gyeongcheol Shin (ORCID: https://orcid.org/0000-0003-4668-7680)
- H.-H.S. Lee
- Changhyun Kim (ORCID: https://orcid.org/0000-0002-3304-4929)
- Wonjun Lee (ORCID: https://orcid.org/0000-0001-5286-6541)
- Seong‐Hun Kim (ORCID: https://orcid.org/0000-0002-5030-1894)
- Gi-Ho Park (ORCID: https://orcid.org/0000-0001-7998-4302)
- Jinkwon Kim
- Dayeon Yun
- Seungyeong Park
- Jongsoon Won (ORCID: https://orcid.org/0009-0001-4632-0072)
Institutions
- Sejong University (KR)
- SK Group (Japan) (JP)
Publication Details
- Published
- 2024-08-25
- DOI
- https://doi.org/10.1109/hcs61935.2024.10664793
- Citations
- 6
- Primary Topic
- Neural Networks and Applications
- Type
- article
- Field-Weighted Citation Impact
- 2.08