Offloading Key Switching on GPUs: A Path towards Seamless Acceleration of FHE

Research output: Contribution to conferencePaperpeer-review

52 Downloads (Pure)

Abstract

Fully Homomorphic Encryption (FHE) enables secure computations on encrypted data, offering strong privacy guarantees for cloud computing, privacy-preserving machine learning, and confidential data processing. However, the computational overhead associated with FHE operations, due to the large size of ciphertext and the high arithmetic complexity, limits its practical applicability.
In this work, we address this challenge by presenting an approach that is implemented within the OpenFHE library in order to offload the most dominant components of key switching for the BGV scheme on GPU hardware. In particular, the scope of this work is the performance improvement of the Approximate Modulus Downscaling (ApproxModDown) function. Our experimental evaluation shows that the proposed system can yield up to a 4.58× performance speedup against the vanilla OpenFHE ApproxModDown implementation, while also resulting in 1.16× performance mprovement per homomorphic multiplication and 1.08× improvement for end-to-end execution time.
Original languageEnglish
Number of pages6
Publication statusAccepted/In press - 7 Apr 2025
EventIEEE International Conference on Cyber Security and Resilience - Chania, Greece
Duration: 4 Aug 20256 Aug 2025
https://www.ieee-csr.org/

Conference

ConferenceIEEE International Conference on Cyber Security and Resilience
Country/TerritoryGreece
CityChania
Period4/08/256/08/25
Internet address

Keywords

  • data privacy
  • fully homomorphic encryption
  • hardware acceleration
  • GPUs

Fingerprint

Dive into the research topics of 'Offloading Key Switching on GPUs: A Path towards Seamless Acceleration of FHE'. Together they form a unique fingerprint.

Cite this