A GPU Microarchitecture Optimized for Fully Homomorphic Encryption

Researchers from Boston University, Northeastern University, KAIST, and University of Murcia, et al. have released “FHECore: Rethinking GPU Microarchitecture for Fully Homomorphic Encryption”.

Abstract
“Fully Homomorphic Encryption (FHE) enables computation directly on encrypted data but incurs massive computational and memory overheads, often exceeding plaintext execution by several orders of magnitude. While custom ASIC accelerators can mitigate these costs, their long time-to-market and the rapid evolution of FHE algorithms threaten their long-term relevance. GPUs, by contrast, offer scalability, programmability, and widespread availability, making them an attractive platform for FHE. However, modern GPUs are increasingly specialized for machine learning workloads, emphasizing low-precision datatypes (e.g., INT8, FP8) that are fundamentally mismatched to the wide-precision modulo arithmetic required by FHE. Essentially, while GPUs offer ample parallelism, their functional units, like Tensor Cores, are not suited for wide-integer modulo arithmetic required by FHE schemes such as CKKS. Despite this constraint, researchers have attempted to map FHE primitives on Tensor Cores by segmenting wide integers into low-precision (INT8) chunks.

To overcome these limitations, the authors propose FHECore, a specialized functional unit integrated directly into the GPU’s Streaming Multiprocessor. The design is motivated by the observation that the two dominant contributors to FHE latency — Number Theoretic Transform and Base Conversion — can be expressed as modulo-linear transformations and mapped onto a common hardware unit that natively supports wide-precision modulo-multiply-accumulate operations. Simulation results demonstrate that FHECore reduces dynamic instruction count by a geometric mean of 2.41× for CKKS primitives and 1.96× for end-to-end workloads. These reductions translate into performance speedups of up to 2.12×, including a 50% reduction in bootstrapping latency, while incurring only a 2.4% area overhead.”

Find the technical paper here. February 2026.

Daksha, Lohit, Seyda Guzelhan, Kaustubh Shivdikar, Carlos Agulló Domingo, Óscar Vera Lopez, Gilbert Jonatan, Hubert Dymarkowski et al. “FHECore: Rethinking GPU Microarchitecture for Fully Homomorphic Encryption.” arXiv preprint arXiv:2602.22229 (2026).