slemesp/GpuFitsCrypt: GpuFitsCrypt v1.0 — GPU-Accelerated AES-GCM Encryption for FITS

Read full text Summarise with VIRAT
Authors: Samuel Lemes-Perera
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Initial public release of GpuFitsCrypt, a high-throughput AES-GCM authenticated encryption library for astronomical FITS catalogs. Features GPU-accelerated AES-GCM with parallel GHASH tree-reduction strategy Dual-stream pipeline for independent header/data encryption Granular access control with separate keys for header and data Policy-driven partial decryption with integrity verification Python wrapper for seamless integration Benchmarked on 7 GPU platforms (Jetson Orin to H100) Validation Bit-exact restoration verified (zero residuals) Scientific validation script with three detection backends (GPUPhot > SEP > scipy) Requirements NVIDIA GPU with CUDA support cfitsio library Python 3.x with astropy, numpy, matplotlib (for examples)

Fields of Study & Subjects

Cryptographic Implementations and Security Parallel Computing and Optimization Techniques Chaos-based Image/Signal Encryption

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
25 March 2026
Language
EN