MNSIM-2.0

Citation Information: Zhenhua Zhu, Hanbo Sun, Tongxin Xie, Yu Zhu, Guohao Dai, Lixue Xia, Dimin Niu, Xiaoming Chen, Xiaobo Sharon Hu, Yu Cao, Yuan Xie, Huazhong Yang, and Yu Wang, MNSIM 2.0: A Behavior-Level Modeling Tool for Processing-In-Memory Architectures, in IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS (TCAD), VOL. 42, NO. 11, NOVEMBER 2023

Main Contributor: Zhenhua Zhu^1*, Hanbo Sun¹, Tongxin Xie¹, Lixue Xia², Gokul Krishnan⁶, Dimin Niu², Qiuwen Lou³,

Xiaoming Chen⁴, Yuan Xie^{2, 5}, Yu Cao⁷, X. Sharon Hu³, Yu Wang^1*, and Huazhong Yang¹

¹Tsinghua University, ²Alibaba Group, ³University of Notre Dame, ⁴Institute of Computing Technology, Chinese Academy of Sciences, ⁵The Hong Kong University of Science and Technology, ⁶Arizona State University ⁷University of Minnesota System

^*[email protected], [email protected]

MNSIM-2.0 aims to model the HW performance and NN computing accuracy of Processing-In-Memory (PIM) architectures. If you have any questions and suggestions about MNSIM-2.0 please contact us via e-mail. We hope that MNSIM-2.0 can be helpful to your research work, and sincerely invite every PIM researcher to add your ideas to MNSIM-2.0 to enlarge its function.

For more information about MNSIM-2.0, please refer to the MNSIM_manual.pdf and the IEEE TCAD paper of MNSIM 2.0.

Weights File: https://1drv.ms/f/s!AtForEDTP2-PgkzYkCJNONO9xnX6?e=y4s0XG