Energy-Efficient Cryogenic Ternary Content Addressable Memory using Ferroelectric SQUID

Abstract

Ternary content addressable memories (TCAMs) are useful for certain computing tasks since they allow us to compare a search query with a whole dataset stored in the memory array. They can also unlock unique advantages for cryogenic applications like quantum computing, high-performance computing, and space exploration by improving speed and energy efficiency through parallel searching. This paper explores the design and implementation of a cryogenic ternary content addressable memory based on ferroelectric superconducting quantum interference devices (FeSQUIDs). The use of FeSQUID for designing the TCAM provides several unique advantages. First, we can get binary decisions (zero or non-zero voltage) for matching and mismatching conditions without using any peripheral circuitry. Moreover, the proposed TCAM needs ultra-low energy (1.36 aJ and 26.5 aJ average energy consumption for 1-bit binary and ternary search, respectively), thanks to the use of energy-efficient SQUIDs. Finally, we show the efficiency of FeSQUID through the brain-inspired application of Hyperdimensional Computing (HDC). Here, the FeSQUID-based TCAM implements the associative memory to support the highly parallel search needed in the inference step. We estimate an energy consumption of 89.4 fJ per vector comparison using a vector size of 10,000 bits. We also compare the FeSQUID-based TCAM array with the 5nm FinFET-based cryogenic SRAM-based TCAM array and observe that the proposed FeSQUID-based TCAM array consumes over one order of magnitude lower energy while performing the same task.

Figures (6)

• Figure 1: Introduction to content addressable memory. Illustrations of the search mechanism with a (a) binary and (b) ternary CAM. (c) Block diagram showing the steps how a CAM can be used for AI applications such as classification and pattern recognition tasks.
• Figure 2: (a) Device structure and circuit symbol of a FeSQUID. (b) Polarization-voltage characteristics of a Led Zirconium Titanate (PZT) ferroelectric material. (c) Current-voltage characteristics of the SQUID for two different polarization states of the ferroelectric material. Device structure and illustration of gate-controlled switching of a hTron channel when (d) $I_G < I_G^C$, keeping the channel in its superconducting state and (e) $I_G > I_G^C$, switching the channel to its resistive state. (f) Gate current-controlled switching of the hTron channel.
• Figure 3: The proposed cryogenic TCAM based on FeSQUID. (a) Circuit schematic of the proposed TCAM cell. (b) Illustration of the array-level organization of the proposed TCAM cells, where data will be stored in a row. (c) Values of RWL currents for two modes of the proposed TCAM. Definitions of the representation of (d) input search data in terms of RBL currents and (e) stored data in terms of FeSQUID's device states.
• Figure 4: Working Principle of the proposed TCAM in the exact search mode. Illustration of the working principle of the TCAM when (a) $Data = 0$ and $Search = 0$, (b)$Data = 0$ and $Search = 1$, (c) $Data = 0$ and $Search = d$, (d) $Data = 1$ and $Search = 0$, (e) $Data = 1$ and $Search = 1$, and (f) $Data = 1$ and $Search = d$. (g) ML voltage, power, and energy consumption for 1-bit exact search with the proposed TCAM.
• Figure 5: ML voltage levels obtained for different number of data lengths and different amounts of mismatch between the input search and stored data.