Description
Date depot: 3 avril 2024
Titre: The DLT Trilemma versus New Adversaries
Directeur de thèse:
Sébastien TIXEUIL (LIP6)
Encadrante :
Silvia BONOMI (Univ. Roma)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Systèmes et réseaux
Resumé: In the last 10 years, Distributed Ledger Technologies (DLTs), and blockchain in particular, have become one of the most widespread technologies to store transactions in a distributed system characterized by full decentralization, transparency, immutability and non-repudiation of data. DLTs represent an example of emerging technology that first consolidated its development, and later had its theoretical foundations investigated. As a consequence, various algorithmic solutions have been proposed, trying to improve the scalability and the resilience to Byzantine (a.k.a. malicious) processes, without sacrificing too much the decentralization of the system (i.e., the DLT trilemma). However, most of the existing solutions lack a solid theoretical analysis formally proving their correctness, and the evaluation is carried out by considering rather static environments, where the system does not change, or changes very slowly mainly due to failures. By contrast, real networks (especially those underlining public permissionless blockchains) are not static and are not free from vulnerabilities exposing them to potential attackers able to progressively compromise multiple
replicas.
Scientific Lock. There still exists an important gap between theoretical studies aimed at defining dependable, robust and scalable algorithms for the core building blocks behind DLTs, and the characteristics of modern practical environments where such communication and computation primitives are actually de-
ployed.