一句话：提出 GPU 高性能 FHE 库 Cheddar，通过 32 位执行、架构定制内核与 kernel fusion 实现 2.18–4.45× 加速。
全同态加密（FHE）允许在加密数据上直接计算，从而缓解云计算的隐私问题，但其计算与内存开销巨大，性能远低于明文处理。为降低开销，作者提出面向 GPU 的高性能 FHE 库 Cheddar。该工作系统性地启用 32 位 FHE 执行，以利用 GPU 内部的 32 位整数数据通路；同时针对特定 GPU 架构优化内核，采用高效底层原语与定制算法。此外，Cheddar 通过调整常见 FHE 操作序列并广泛使用 kernel fusion，缓解内存带宽压力。在代表性 FHE 工作负载上，Cheddar 相比当前最优 GPU 实现取得 2.18–4.45× 的性能提升。
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× for representative FHE workloads compared to state-of-the-art GPU implementations.