An LPDDR-based CXL-PNM Platform for TCO-efficient Inference of Transformer-based Large Language Models

التفاصيل البيبلوغرافية
العنوان: An LPDDR-based CXL-PNM Platform for TCO-efficient Inference of Transformer-based Large Language Models
المؤلفون: Park, Sang-Soo, Kim, KyungSoo, So, Jinin, Jung, Jin, Lee, Jonggeon, Woo, Kyoungwan, Kim, Nayeon, Lee, Younghyun, Kim, Hyungyo, Kwon, Yongsuk, Kim, Jinhyun, Lee, Jieun, Cho, YeonGon, Tai, Yongmin, Cho, Jeonghyeon, Song, Hoyoung, Ahn, Jung Ho, Kim, Nam Sung
المصدر: 2024 IEEE International Symposium on High-Performance Computer Architecture (HPCA) HPCA High-Performance Computer Architecture (HPCA), 2024 IEEE International Symposium on. :970-982 Mar, 2024
Relation: 2024 IEEE International Symposium on High-Performance Computer Architecture (HPCA)
قاعدة البيانات: IEEE Xplore Digital Library