TAP-CAM: A Tunable Approximate Matching Engine based on Ferroelectric Content Addressable Memory

TL;DR

This work introduces TAP-CAM, a tunable approximate matching engine built on a 2FeFET-2R TCAM cell with an evaluation transistor to realize bit-by-bit Hamming-distance thresholds. By combining compact FeFET-based storage with controllable ML discharge, TAP-CAM achieves precise threshold matching while maintaining low energy consumption. The approach demonstrates substantial energy savings and accuracy gains in KNN benchmarks compared with exact-match CMOS CAM and improved efficiency versus prior approximate CAM designs. Overall, TAP-CAM offers a practical, scalable solution for energy-efficient, tunable approximate searching in memory-centric data-processing workloads.

Abstract

Pattern search is crucial in numerous analytic applications for retrieving data entries akin to the query. Content Addressable Memories (CAMs), an in-memory computing fabric, directly compare input queries with stored entries through embedded comparison logic, facilitating fast parallel pattern search in memory. While conventional CAM designs offer exact match functionality, they are inadequate for meeting the approximate search needs of emerging data-intensive applications. Some recent CAM designs propose approximate matching functions, but they face limitations such as excessively large cell area or the inability to precisely control the degree of approximation. In this paper, we propose TAP-CAM, a novel ferroelectric field effect transistor (FeFET) based ternary CAM (TCAM) capable of both exact and tunable approximate matching. TAP-CAM employs a compact 2FeFET-2R cell structure as the entry storage unit, and similarities in Hamming distances between input queries and stored entries are measured using an evaluation transistor associated with the matchline of CAM array. The operation, robustness and performance of the proposed design at array level have been discussed and evaluated, respectively. We conduct a case study of K-nearest neighbor (KNN) search to benchmark the proposed TAP-CAM at application level. Results demonstrate that compared to 16T CMOS CAM with exact match functionality, TAP-CAM achieves a 16.95x energy improvement, along with a 3.06% accuracy enhancement. Compared to 2FeFET TCAM with approximate match functionality, TAP-CAM achieves a 6.78x energy improvement.

Figures (10)

• Figure 1: (a) FeFET polarization directions and channel conditions after memory write operations; (b) The FeFET $\textit{I}_\textit{D}$-$\textit{V}_\textit{G}$ characteristics after positive/negative gate write; (c) 1FeFET-1R structure and equivalent circuit; (d) The 1FeFET-1R $\textit{I}_\textit{D}$-$\textit{V}_\textit{G}$ characteristics after positive/negative gate write.
• Figure 2: Schematics of (a) 16T CMOS TCAM cell; (b) 2T-2ReRAM TCAM cell; (c) 20T-6MTJ TCAM cell; (d) 2FeFET TCAM cell.
• Figure 3: (a) Exact match: The stored entry that matches exactly with the query; (b) Best match: The stored entry that has the smallest distance to the query; (c) Threshold match: The stored entry whose distance to the query is below specified thresholds.
• Figure 4: (a) Structure of the proposed 2FeFET-2R TCAM cell; (b) Transient voltage waveforms of 2FeFET-2R CAM cell storing '1’.
• Figure 5: Structure of a 2FeFET-2R TCAM array with wordlength 64.