Fully Homomorphic Encryption (FHE) relies heavily on the Number Theoretic Transform (NTT), making NTT a major performance bottleneck due to its intensive polynomial computations. Hybrid Homomorphic Encryption (HHE), which integrates arithmetic and logic FHE, further requires support for multiple NTT lengths. However, existing accelerators mainly optimize NTT throughput and do not provide unified support for HHE. This paper presents Hermes, a unified high-performance NTT architecture based on hybrid dataflow. Hermes exploits parallelism along both temporal and spatial dimensions and incorporates a fully pipelined on-chip computing core. A conflict-free on-chip fragmentation algorithm is introduced to resolve bank conflicts and enable burst HBM access, while an efficient dataflow improves computational intensity through data reuse, reducing bandwidth demand. Experimental results show that Hermes supports multiple NTT lengths and achieves up to 13.6x and 1.3x higher throughput than state-of-the-art GPU and FPGA accelerators, respectively. Our source code is available at https://anonymous.4open.science/r/Hermes_conf-4E6F.

翻译：全同态加密（FHE）高度依赖于数论变换（NTT），而NTT因其密集的多项式计算成为主要性能瓶颈。混合同态加密（HHE）融合了算术与逻辑FHE，进一步要求支持多种NTT长度。然而，现有加速器主要优化NTT吞吐量，并未为HHE提供统一支持。本文提出Hermes，一种基于混合数据流的高性能统一NTT架构。Hermes同时利用时间和空间维度的并行性，并集成了全流水化的片上计算核心。通过引入无冲突的片上分块算法来解决存储体冲突并实现突发式HBM访问，同时采用高效数据流通过数据复用提升计算强度，从而降低带宽需求。实验结果表明，Hermes支持多种NTT长度，其吞吐量分别达到最先进GPU和FPGA加速器的13.6倍和1.3倍。我们的源代码公开于 https://anonymous.4open.science/r/Hermes_conf-4E6F。