Abstract
dc:description.abstractThe architecture of the Web was designed to enable decentralised exchange of information. Early architects envisioned an egalitarian yet organic society thriving in cyberspace. The reality of the Web today, unfortunately, does not bear out these visions: information networks have repeatedly shown a tendency towards consolidation and centralisation with the current Web split between a handful of large corporations. The advent of Bitcoin and successor blockchain networks re-ignited interest in developing alternatives to the centralised Web and paving a way back to the earlier architectural visions for the Web. This has led to immense hype around these technologies with the cryptocurrency market valued at several hundred billions of dollars at the time of writing. With great hype, apparently, come great scams. I start off by analysing the use of Bitcoin as an enabler for crime and then present both technical solutions as well as policy recommendations to mitigate the harm these crimes cause. These policy recommendations then lead us on to look more closely at cryptocurrency's tamer cousin: permissioned blockchains. These systems, while less revolutionary in their premise, nevertheless aim to provide sweeping improvements in the efficiency and transparency of existing enterprise systems. To see whether they work in practice, I present the results of my work in delivering a production permissioned blockchain system to real users. This involves comparing several permissioned blockchain systems, exploring their deficiencies and developing solutions for the most egregious of those. Lastly, I do a deep dive into one of the most persistent technical issues with permissioned blockchains, and decentralised networks in general: the lack of scalability in their consensus mechanisms. I present two novel consensus algorithms that aim to improve upon the state of the art in several ways. The first is designed to enable existing permissioned blockchain networks to scale to thousands of nodes. The second presents an entirely new way of building decentralised consensus systems utilising a trie-based data structure at its core as opposed to the usual linear ledgers used in current systems.
Degree
thesis:*- Name dc:type.qualificationname
- Doctor of Philosophy (PhD)
- Level dc:type.qualificationlevel
- Doctoral
- Grantor dc:publisher.institution
- University of Cambridge
- Year dc:date.issued
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ahmed, Mansoor
- Advisor dc:contributor.advisor
-
- Anderson, Ross
Subjects
dc:subject × 3Rights
dc:rightsIdentifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.17863/CAM.74213
- OAI identifier oai:identifier
- oai:www.repository.cam.ac.uk:1810/326765