Description

This thesis aims to advance scientific knowledge in decentralized data storage by combining algorithm design, data availability verification, and rigorous validation of the proposed solutions.

The first objective is to develop new distributed algorithms that emphasize algorithmic frugality, in order to reduce computational, memory, and energy requirements. This approach helps lower infrastructure costs, facilitates deployment in resource-constrained environments, and contributes to more sustainable computing.

The second objective focuses on designing mechanisms to efficiently prove the actual possession of data in architectures distributed across multiple independent nodes. The work will target lightweight, scalable protocols capable of verifying data availability without massive data transfers or network overhead.

Finally, the third objective concerns the formal verification of the developed algorithms and protocols to ensure their correctness and reliability. Mathematical modeling and automated analysis will strengthen operational robustness, scientific reproducibility, and the potential for transfer to industrial applications.

Skills required

Rigorous, with a solid foundation in core computer science. Comfortable with prototyping for proof-of-concept development.

Bibliography

Pascal Lafourcade, Lola-Baie Mallordy, Charles Olivier-Anclin, Léo Robert:
Secure Keyless Multi-party Storage Scheme. ESORICS (3) 2024: 279-298

Gaël Marcadet, Pascal Lafourcade, Léo Robert:
RMC-PVC: A Multi-Client Reusable Verifiable Computation Protocol. SAC 2023: 1558-1565

Pascal Lafourcade, Gaël Marcadet, Léo Robert:
iUC-Secure Distributed File Transfer from Standard Attribute-Based Encryption. AFRICACRYPT 2024: 174-198

Gil Utard:
Calcul sur les données volumineuses et stockage distribué à grande échelle. University of Picardie Jules Verne, Amiens, France, 2023

Keywords

Access control, data integrity, provable security, data storage