PE-AONT (Partial Encryption All or Nothing Transform) represents the most recent iteration (2022) of an algorithm designed to fragment and disperse data over different independent servers or storage sites. It combines encryption and non-encryption techniques to achieve exposure protection and faster processing. However, these claims have yet to be confirmed, partly due to the lack of publicly available source code. In this work, we present the first publicly available Java reimplementation of PE-AONT, built on Docker containers for reproducibility, and evaluate it on an IoT network comprising five Raspberry Pi nodes. In addition to providing evidence of markedly different performance characteristics in our setup ((Formula presented) vs. the claimed (Formula presented) ), we extend the scheme by adding a previously missing data reconstruction service and an exhaustive performance characterization across a range of configurations. We release the full PE-AONT implementation in the open-science spirit.
Reimplementation and performance characterization of the PE-AONT schema for efficient security and retrieval of confidential data at the edge
Casciaro, Simone
;Mainetti, Luca
2026-01-01
Abstract
PE-AONT (Partial Encryption All or Nothing Transform) represents the most recent iteration (2022) of an algorithm designed to fragment and disperse data over different independent servers or storage sites. It combines encryption and non-encryption techniques to achieve exposure protection and faster processing. However, these claims have yet to be confirmed, partly due to the lack of publicly available source code. In this work, we present the first publicly available Java reimplementation of PE-AONT, built on Docker containers for reproducibility, and evaluate it on an IoT network comprising five Raspberry Pi nodes. In addition to providing evidence of markedly different performance characteristics in our setup ((Formula presented) vs. the claimed (Formula presented) ), we extend the scheme by adding a previously missing data reconstruction service and an exhaustive performance characterization across a range of configurations. We release the full PE-AONT implementation in the open-science spirit.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


