A Computing-in-Memory Engine for Searching on Homomorphically Encrypted Data

Abstract

The high volumes of data stored in the cloud, coupled with growing concerns about security and privacy, have motivated research on homomorphic encryption (HE), i.e., a technique that enables computation directly on encrypted data, obviating the need for prior decryption. Recent algorithmic advances have enabled a diverse set of homomorphic operations (e.g., addition, multiplication, and division). Looking at the applications, recent work also suggests extensibility secure, homomorphically encrypted content-addressable memories [or secure content-addressable memories (SCAMs)]. Still, the large datawords that result from homomorphic data encodings (i.e., that must be stored/transferred for computation), compounded with the implicit computational complexity of HE, still impede the deployment of homomorphic computer hardware. As an alternative, computing-in-memory (CiM) architectures could significantly reduce the volume of data transfers for SCAM (and other) applications, leading to considerable energy savings and latency reduction. In this regard, we propose a CiM-compatible engine for SCAM (CiM-SCAM) and analyze the pros and cons of three different memory cells: a 6T CMOS SRAM and two memory cells based on ferroelectric field-effect transistors (FeFETs) (specifically 2T + 1FeFET and 1-FeFET designs). CiM-SCAM leverages in-place copy buffers (IPCBs), along with customized sense amplifiers that include two types of in-memory adders. Our results suggest that energy (and search time) improvements of up to 16× (3.2×) for 1-FeFET memory cells are possible, compared with an application-specific integrated circuit (ASIC) approach. Similar improvements are also possible with SRAM and 2T + 1-FeFET memory cells. For the latter, we achieve up to 13× (3.1×) of energy (speedup). INDEX TERMS Computing-in-memory (CiM), emerging technologies, ferroelectric field-effect transistors (FeFETs), homomorphic encryption (HE), secure content-addressable memory (CAM), SRAMs.

Highly Robust Flexible Ferroelectric Field Effect Transistors Operable at High Temperature with Low‐Power Consumption

“…In fact, there is no previous flexible FeFETs reported to possess high temperature robustness as we have demonstrated here. The device of course has to be integrated into arrays with multiple cells for memory applications, and we expect silicon‐based processing and architecture can be translated into mica‐based platform.…”

The Implications of Ferroelectric FET Device Models to the Design of Computing-in-Memory Architectures

“…The FeFET-CiM architecture based on 2T+1FeFETs offers superior application-level benefits when compared to the same architecture based on 1Fe-FETs. Nevertheless, CiM architectures based on 1FeFET memory cells may offer up to ∼5.3X memory density improvements when compared to conventional 6T-SRAM architectures [15,14]. Finally, 1FeFET-CiM still offers relevant application-level improvements when compared to a counterpart STT-CiM architecture.…”

“…The high density of the 1FeFET memory cells is another advantage of the 1-FeFET-CiM architecture. As discussed in [14,15], arrays based on 1FeFET memory cells may offer up to ∼5.3X density improvements when compared to conventional 6T-SRAM arrays. The high density and low leakage power (that arises from non-volatility) make 1FeFET memory cells great candidates for the design of low-leakage and dense CiM architectures.…”

“…IV., the FeFET-CiM architecture based on 2T+1FeFETs (1FeFETs) offers an average speedup of ∼2.5X (∼1.1X) and energy reduction of ∼1.7X (∼1.4X) when compared to a SRAM baseline across a wide range of benchmark programs from different domains. Despite smaller speedups and energy savings when compared to 2T+1FeFET-CiM, 1FeFET memory arrays may offer up to ∼5.3X density improvements when compared to conventional 6T-SRAM arrays [14,15]. Furthermore, 1FeFET-CiM offers significant application-level improvements when compared to a counterpart STT-CiM architecture.…”

IMCRYPTO: An In-Memory Computing Fabric for AES Encryption and Decryption

“…4(a)). The CEM of IMCRYPTO employ circuits similar to those in [15], [23] -i.e., customized sense amplifiers -which enable bitwise Boolean logic between two memory words stored in a memory array.…”