MN-Core - A Highly Efficient and Scalable Approach to Deep Learning

K. Namura,Johannes Maximilian Kühn, Tohru Adachi, H. Imachi, H. Kaneko,T. Kato, Go Watanabe, Naoto Tanaka, S. Kashihara, Hiroshi Miyashita, Y. Tomonaga,Ryosuke Okuta,Takuya Akiba,Brian Vogel, S. Kitajo, F. Osawa,K. Takahashi, Y. Takatsukasa, K. Mizumaru, T. Yamauchi, J. Ono,A. Takahashi,Tanvir Ahmed, Y. Doi, K. Hiraki, J. Makino

2021 Symposium on VLSI Circuits（2021）

引用 1|浏览16

暂无评分

摘要

MN-Core is a highly efficient deep learning training accelerator reaching in excess of 1 TFLOPS/W (half-precision) at board level in real-world mixed-precision workloads. To reach and sustain this level of performance, the design is partitioned and packaged as four-die MCM package exceeding 3000mm ² of die area.

查看译文

关键词

Accelerator,MCM,Deep Learning,HPC,SIMD

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要