EUVIMEDEUVIMED entdecken
Crossref

TensorFHE+: Fully Homomorphic Encryption Acceleration Based on Linear Algebra

Yintai Sun, Shengyu Fan, Zhenhua Yin, Xinkai Song, Xing Hu, Zidong Du, Qi Guo, Weizhi Xu, Rui Hou, Dan Meng, Song Bian, Mingzhe Zhang

IEEE Transactions on Computers · 2026-02 · 75 (2) · 612-627

DOI: 10.1109/tc.2025.3629614 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Computers, 2026-02. Verfasst von Yintai Sun und 11 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 45 Literaturangaben. Gefördert von: National Key R&D Program of China, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fully Homomorphic Encryption (FHE) enables encrypted data processing on untrusted cloud servers, crucial for privacy-sensitive applications. Despite its potential, performance overheads (about $10,000\times$ 10 , 000 × slower) limit adoption. ASIC accelerators outperform GPUs/FPGAs by optimizing specific operations but rely on costly 7nm processes and large on-chip memory, hindering cost-effective deployment. Balancing efficiency with manufacturing constraints remains critical. This paper presents TensorFHE+, a GPU-optimized FHE acceleration framework leveraging Tensor Cores to accelerate Number Theoretic Transform (NTT) operations. Key innovations include: 1) Decomposing CKKS kernels into vector/matrix operations for hardware utilization; 2) Vectorized modulo arithmetic; 3) Data layout optimization for memory efficiency. Evaluated on NVIDIA A100, TensorFHE+ outperforms TensorFHE[1] by $1.44\times$ 1.44 × in average (up to $1.69\times$ 1.69 × on ResNet-20) and surpasses prior GPU implementations [2] , [3] . The design also demonstrates compatibility with commercial linear algebra accelerators, enabling efficient FHE deployment.

Förderung

  • National Key R&D Program of China
  • National Natural Science Foundation of China

Zitieren

APA

Sun, Y., Fan, S., Yin, Z., Song, X., Hu, X., Du, Z., Guo, Q., Xu, W., Hou, R., Meng, D., Bian, S., & Zhang, M. (2026). TensorFHE+: Fully Homomorphic Encryption Acceleration Based on Linear Algebra. IEEE Transactions on Computers, 75(2), 612-627. https://doi.org/10.1109/tc.2025.3629614

Vancouver

Sun Y, Fan S, Yin Z, Song X, Hu X, Du Z, et al. TensorFHE+: Fully Homomorphic Encryption Acceleration Based on Linear Algebra. IEEE Transactions on Computers. 2026;75(2):612-627. doi:10.1109/tc.2025.3629614

BibTeX
@article{Sun2026,
  author = {Sun, Yintai and Fan, Shengyu and Yin, Zhenhua and Song, Xinkai and Hu, Xing and Du, Zidong and Guo, Qi and Xu, Weizhi and Hou, Rui and Meng, Dan and Bian, Song and Zhang, Mingzhe},
  title = {TensorFHE+: Fully Homomorphic Encryption Acceleration Based on Linear Algebra},
  journal = {IEEE Transactions on Computers},
  year = {2026},
  volume = {75},
  number = {2},
  pages = {612--627},
  doi = {10.1109/tc.2025.3629614}
}
Kostenlos starten →