In-Memory Nearest Neighbor Search With Nanoelectromechanical Ternary Content-Addressable Memory
- 29 November 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Electron Device Letters
- Vol. 43 (1) , 154-157
Abstract
Nearest neighbor (NN) search is widely used in pattern classification and memory-augmented neural networks. To overcome the von Neumann bottleneck in conventional NN search architecture, in this study, nanoelectromechanical-switch-based ternary content-addressable memory (NEMTCAM) is introduced for the NN classifier. NEMTCAM can calculate the Hamming distance between the input vector and the stored vectors in a parallel search operation. The NEMTCAM operation was experimentally demonstrated. Furthermore, an analytical model for NN search accuracy, including cell-to-cell parasitic resistance, is presented. NEMTCAM can calculate up to 10 Hamming distances in 32-bit words owing to the high current ratio of the NEM memory switches.Keywords
Funding Information
- National Research Foundation (NRF) of Korea funded by the Ministry of Science and ICT (MSIT) through the Intelligent Semiconductor Technology Development Program (NRF-2019M3F3A1A02072089, NRF-2021M3F3A2A01037927, NRF-2021R1A2C1007931)
- Institute for Information and Communications Technology Planning and Evaluation (IITP) funded by the MSIT through the Information Technology Research Center Program (IITP-2020-2018-0-01421)
- Ministry of Trade, Industry and Energy/Korea Semiconductor Research Consortium (MOTIE/KSRC) through the Technology Innovation Program (10080575)
This publication has 10 references indexed in Scilit:
- Nanoelectromechanical-Switch-Based Ternary Content-Addressable Memory (NEMTCAM)IEEE Transactions on Electron Devices, 2021
- Tri-State Nanoelectromechanical Memory Switches for the Implementation of a High-Impedance StateIEEE Access, 2020
- Parasitic Resistance Effect Analysis in RRAM-based TCAM for Memory Augmented Neural Networks2020
- Ferroelectric ternary content-addressable memory for one-shot learningNature Electronics, 2019
- In-depth Characterization of Resistive Memory-Based Ternary Content Addressable Memories2018
- Write Disturb in Ferroelectric FETs and Its Implication for 1T-FeFET AND Memory ArraysIEEE Electron Device Letters, 2018
- NNgine: Ultra-Efficient Nearest Neighbor Accelerator Based on In-Memory Computing2017
- Monolithic Three-Dimensional 65-nm CMOS-Nanoelectromechanical Reconfigurable Logic for Sub-1.2-V OperationIEEE Electron Device Letters, 2017
- Clustering Billions of Images with Large Scale Nearest Neighbor SearchIEEE Winter Conference on Applications of Computer Vision (WACV), 2007
- Nearest neighbor pattern classificationIEEE Transactions on Information Theory, 1967