Comparative Evaluation of Memory Technologies for Synaptic Crossbar Arrays -- Part I: Robustness-driven Device-Circuit Co-Design and System Implications
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929457943543808 |
|---|---|
| author | Wang, Chunguang Victor, Jeffry Gupta, Sumeet K. |
| author_facet | Wang, Chunguang Victor, Jeffry Gupta, Sumeet K. |
| contents | In-memory computing (IMC) utilizing synaptic crossbar arrays is promising for energy-efficient deep neural network (DNN) accelerators. Various technologies (CMOS and post-CMOS) have been explored as synaptic device candidates, each with its own pros and cons. In this work, we perform a design space exploration and comparative evaluation of four technologies viz. 8T SRAMs, ferroelectric transistors (FeFETs), resistive RAMs (ReRAMs) and spin-orbit torque magnetic RAMs (SOT-MRAMs) in the context of IMC robustness and DNN accuracy. For a fair comparison, we carefully optimize each technology specifically for synaptic crossbar design accounting for device and circuit non-idealities. By integrating different technologies into a cross-layer simulation flow based on physical models of synaptic devices and interconnects, we present insights into various device-circuit interactions. Based on the optimized designs, we obtain inference results for ResNet-20 on CIFAR-10 dataset. Among the four technologies, we show that FeFETs-based DNNs achieve the highest accuracy (followed closely by ReRAMs) and the largest resilience to process variations due to the compactness and high ON/OFF current ratio of FeFET bit-cells. In Part II of this paper, we expand the technology evaluation considering various device-circuit design knobs used for crossbar arrays. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_04261 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Comparative Evaluation of Memory Technologies for Synaptic Crossbar Arrays -- Part I: Robustness-driven Device-Circuit Co-Design and System Implications Wang, Chunguang Victor, Jeffry Gupta, Sumeet K. Emerging Technologies In-memory computing (IMC) utilizing synaptic crossbar arrays is promising for energy-efficient deep neural network (DNN) accelerators. Various technologies (CMOS and post-CMOS) have been explored as synaptic device candidates, each with its own pros and cons. In this work, we perform a design space exploration and comparative evaluation of four technologies viz. 8T SRAMs, ferroelectric transistors (FeFETs), resistive RAMs (ReRAMs) and spin-orbit torque magnetic RAMs (SOT-MRAMs) in the context of IMC robustness and DNN accuracy. For a fair comparison, we carefully optimize each technology specifically for synaptic crossbar design accounting for device and circuit non-idealities. By integrating different technologies into a cross-layer simulation flow based on physical models of synaptic devices and interconnects, we present insights into various device-circuit interactions. Based on the optimized designs, we obtain inference results for ResNet-20 on CIFAR-10 dataset. Among the four technologies, we show that FeFETs-based DNNs achieve the highest accuracy (followed closely by ReRAMs) and the largest resilience to process variations due to the compactness and high ON/OFF current ratio of FeFET bit-cells. In Part II of this paper, we expand the technology evaluation considering various device-circuit design knobs used for crossbar arrays. |
| title | Comparative Evaluation of Memory Technologies for Synaptic Crossbar Arrays -- Part I: Robustness-driven Device-Circuit Co-Design and System Implications |
| topic | Emerging Technologies |
| url | https://arxiv.org/abs/2307.04261 |