CMAS Lab - Impact of Scaling on Large Crossbar Arrays for Neuromorphic

Journal article

Suyash Kushwaha, Chintu Bhaskara Rao, Shamini P R, Sourajeet Roy, Rohit Sharma
Workshop on Signal Propagation on Interconnects, 2025

Semantic Scholar DOI

Cite

APA Click to copy
Kushwaha, S., Rao, C. B., R, S. P., Roy, S., & Sharma, R. (2025). Impact of Scaling on Large Crossbar Arrays for Neuromorphic Applications. Workshop on Signal Propagation on Interconnects.

Chicago/Turabian Click to copy
Kushwaha, Suyash, Chintu Bhaskara Rao, Shamini P R, Sourajeet Roy, and Rohit Sharma. “Impact of Scaling on Large Crossbar Arrays for Neuromorphic Applications.” Workshop on Signal Propagation on Interconnects (2025).

MLA Click to copy
Kushwaha, Suyash, et al. “Impact of Scaling on Large Crossbar Arrays for Neuromorphic Applications.” Workshop on Signal Propagation on Interconnects, 2025.

BibTeX Click to copy

@article{suyash2025a,
  title = {Impact of Scaling on Large Crossbar Arrays for Neuromorphic Applications},
  year = {2025},
  journal = {Workshop on Signal Propagation on Interconnects},
  author = {Kushwaha, Suyash and Rao, Chintu Bhaskara and R, Shamini P and Roy, Sourajeet and Sharma, Rohit}
}

Abstract

Large-scale crossbar arrays form the functional building block for neural networks that perform efficient inmemory computing with massive parallelism. In this paper, the impact of scaling of nano interconnects in large-sized crossbar arrays for neuromorphic applications is presented. Nanoscale interconnects are used to generate significantly larger crossbar arrays like $64 \times 64,128 \times 128$ and $256 \times 256$ considering 13 nm and 7 nm technology nodes. These crossbar arrays are analyzed to study the effects of technology scaling and large arrays sizing on the key performance metrics like delay time, rise time and fall time.