Back to search

Cornell University

MINIMIZING TRUST IN DECENTRALIZED SYSTEMS

Abstract

dc:description.abstract

Modern distributed systems typically optimize for ease of deployment and performance. However, this comes at the cost of trusting a single entity to control all aspects of the system. While this has allowed for the proliferation of computational power in modern life, the shortcomings of such a compromise are becoming increasingly apparent. Having a single administrative point of failure leaves these systems vulnerable to data breaches or undesired actions being taken by the administrator. In the last decade, Bitcoin demonstrated the viability of a system where no single entity has unilateral control. While the lastdecade has proven the promise of decentralized systems, there are still many axes where such systems remain woefully centralized. Much of the work done to combat centralizing forces is specific to a particular system or implementation. This thesis finds general purpose solutions to remove points of centralization from large classes of decentralized systems irrespective of their underlying architecture. First, it introduces a novel method to measure some of the underlying causes of centralization in proof of work cryptocurrencies. Next, it removes the ability for any replica to unilaterally divulge sensitive information through a mechanism that efficiently handles secret shared state in a permissioned blockchain. Then, this thesis presents a new transaction fee mechanism that inhibits the ability for a replica to have undue influence over the transaction order. Finally, this thesis illustrates how to apply some of these principles in a more traditionally centralized setting by reducing the amount of trust placed in IoT hubs in IoT deployments.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Computer Science
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Computer Science
Grantor
Cornell University
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Basu, Soumya
Chair dc:contributor.chair
  • Sirer, Emin
Committee members dc:contributor.committeemember
  • Agarwal, Rachit
  • O'Hara, Maureen

Subjects

dc:subject × 6

Rights

dc:rights
Statement dc:rights
  • Attribution 4.0 International
Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Bibid
16938381
Dc Identifier Other
ProQuest Submission ID: 14835
ProQuest Publication ID: 31840735

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Basu, Soumya. MINIMIZING TRUST IN DECENTRALIZED SYSTEMS. Doctor of Philosophy thesis, Cornell University, 2025. https://hdl.handle.net/1813/117534