2025 IEEE International Conference on Cyber Security and Resilience

Full Program

Summary:

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$\times$ performance speedup against the vanilla OpenFHE ApproxModDown implementation, while also resulting in
1.16$\times$ performance improvement per homomorphic multiplication and 1.08$\times$ improvement for end-to-end execution time.

Author(s):

Orion Papadakis    
United Kingdom

Michail Papadimitriou    
United Kingdom

Athanasios Stratikopoulos    
United Kingdom

Maria Xekalaki    
United Kingdom

Juan Fumero    
United Kingdom

Christos Kotselidis    
United Kingdom

 


Copyright © 2025 SUMMIT-TEC GROUP LTD