Cheddar: A Swift Fully Homomorphic Encryption Library Designed for GPU Architectures

TL;DR

Cheddar is a GPU-optimized fully homomorphic encryption library that significantly accelerates encrypted computations by leveraging 32-bit integer operations, kernel optimizations, and memory bandwidth improvements.

Contribution

The paper introduces Cheddar, a novel GPU-based FHE library that achieves substantial speedups through architecture-specific optimizations and kernel fusion techniques.

Findings

Achieves 2.18--4.45× speedup over previous GPU FHE implementations

Enables 32-bit FHE execution on GPUs using optimized kernels

Reduces memory bandwidth bottleneck through operational adjustments

Abstract

Fully homomorphic encryption (FHE) frees cloud computing from privacy concerns by enabling secure computation on encrypted data. However, its substantial computational and memory overhead results in significantly slower performance compared to unencrypted processing. To mitigate this overhead, we present Cheddar, a high-performance FHE library for GPUs, achieving substantial speedups over previous GPU implementations. We systematically enable 32-bit FHE execution, leveraging the 32-bit integer datapath within GPUs. We optimize GPU kernels using efficient low-level primitives and algorithms tailored to specific GPU architectures. Further, we alleviate the memory bandwidth burden by adjusting common FHE operational sequences and extensively applying kernel fusion. Cheddar delivers performance improvements of 2.18--4.45$\times$ for representative FHE workloads compared to state-of-the-art GPU implementations.