Trinity: A General Purpose FHE Accelerator

TL;DR

Trinity is a versatile FHE accelerator that efficiently supports multiple schemes like CKKS and TFHE within a unified architecture, achieving significant performance gains with minimal hardware overhead.

Contribution

We introduce Trinity, the first multi-modal FHE accelerator that unifies support for CKKS, TFHE, and their conversion, optimizing hardware utilization and performance.

Findings

Outperforms state-of-the-art accelerators by 1.49x for CKKS and 4.23x for TFHE.

Achieves 919.3x performance improvement over CPU-based implementations.

Maintains high hardware utilization with only 85% of the combined circuit area overhead.

Abstract

In this paper, we present the first multi-modal FHE accelerator based on a unified architecture, which efficiently supports CKKS, TFHE, and their conversion scheme within a single accelerator. To achieve this goal, we first analyze the theoretical foundations of the aforementioned schemes and highlight their composition from a finite number of arithmetic kernels. Then, we investigate the challenges for efficiently supporting these kernels within a unified architecture, which include 1) concurrent support for NTT and FFT, 2) maintaining high hardware utilization across various polynomial lengths, and 3) ensuring consistent performance across diverse arithmetic kernels. To tackle these challenges, we propose a novel FHE accelerator named Trinity, which incorporates algorithm optimizations, hardware component reuse, and dynamic workload scheduling to enhance the acceleration of CKKS, TFHE, and their conversion scheme. By adaptive select the proper allocation of components for NTT and MAC, Trinity maintains high utilization across NTTs with various polynomial lengths and imbalanced arithmetic workloads. The experiment results show that, for the pure CKKS and TFHE workloads, the performance of our Trinity outperforms the state-of-the-art accelerator for CKKS (SHARP) and TFHE (Morphling) by 1.49x and 4.23x, respectively. Moreover, Trinity achieves 919.3x performance improvement for the FHE-conversion scheme over the CPU-based implementation. Notably, despite the performance improvement, the hardware overhead of Trinity is only 85% of the summed circuit areas of SHARP and Morphling.