Welcome to HEonGPU’s documentation!

HEonGPU is a high-performance, open-source C++ library designed to accelerate Fully Homomorphic Encryption (FHE) operations by leveraging the immense parallel processing capabilities of modern Graphics Processing Units (GPUs).

This documentation provides a comprehensive guide to the library, from installation and basic usage to deep technical details and API references.

Contents:

• Introduction to HEonGPU
  • Project Goals and Philosophy
  • Key Features and Capabilities
  • The HEonGPU Ecosystem
• Getting Started
  • Prerequisites
  • Building the Library
  • Verifying the Installation
  • Your First Encrypted Computation
• User Guide
  • Core Library Concepts
  • Parameter Selection Guide
  • Storage Management with ExecutionOptions
  • Serialization
  • Tutorials
• Technical Deep Dive
  • Library Architecture
  • Accelerating Polynomial Arithmetic: The Number Theoretic Transform (NTT)
  • Memory Management with RAPIDS Memory Manager (RMM)
  • Decomposition of Homomorphic Operations
• Bootstrapping in HEonGPU
  • CKKS Bootstrapping
  • TFHE Bootstrapping
• Library Examples
  • Basic Operations
  • Advanced Features
  • Serialization
  • Bootstrapping Examples
  • Multi-Party Computation Examples
• API Reference
  • Core Enumerations
  • Context and Parameter Classes
  • Key Management Classes
  • Data and Cryptographic Classes
• Full API Listing
  • Full API
• Advanced Topics
  • Configuration (defines.h)
  • Multiparty Computation (MPC)
  • Using HEonGPU in a Downstream CMake Project
  • Project Roadmap
• Contributing to HEonGPU
  • How You Can Contribute
  • License
• Appendix
  • Performance Benchmarks
  • How to Cite HEonGPU

Indices and tables

• Index

• Search Page