TMD-TCAM: Transition Metal Dichalcogenide Ternary Content Addressable Memory Enables Massive Parallel Data Search

  • Authors:
    Rui Yang (Stanford), Haitong Li (Stanford), Kirby Smithe (Stanford), Taeho Kim (Stanford), Kye Okabe (Stanford), Eric Pop (Stanford), Jonathan Fan (Stanford), H.-S. Philip Wong (Stanford)
    Publication ID:
    P094160
    Publication Type:
    Paper
    Received Date:
    3-Aug-2018
    Last Edit Date:
    9-Jan-2019
    Research:
    2776.003 (Stanford University)
    2776.040 (Stanford University)

Abstract

Ternary content-addressable memory (TCAM) is specialized hardware for in-memory search and pattern-matching for data-intensive applications. However, challenges remain in achieving TCAMs with high search capacity, good area- and energy-efficiency. In this work, the researchers integrate two-dimensional (2D) MoS2 transistors with resistive random-access memories (RRAMs) into 2-transistor-2-resistor (2T2R) TCAM cells for the first time. The MoS2 transistors have very low leakage current and can program the RRAMs with exceptionally robust current control, enabling very large number of bits for parallel search. These TCAM cells show remarkably large resistance ratio (R-ratio) between match and mismatch states, up to 8.5×105. This R-ratio is comparable to that of commercial TCAMs using static random-access memories (SRAMs), with the key advantage that the 2T2R TCAMs use far fewer transistors and have zero stand-by power due to RRAM’s non-volatility. These results also represent a key application of 2D transistors, taking advantage of their high performance yet low leakage.

4819 Emperor Blvd, Suite 300 Durham, NC 27703 Voice: (919) 941-9400 Fax: (919) 941-9450

Important Information for the SRC website. This site uses cookies to store information on your computer. By continuing to use our site, you consent to our cookies. If you are not happy with the use of these cookies, please review our Cookie Policy to learn how they can be disabled. By disabling cookies, some features of the site will not work.