Abstract
dc:description.abstractIn the last two decades, two significant technological developments were the advent of cryptocurrencies and the onset of the cloud computing era. Cryptocurrencies introduced new ways to think about decentralized network cooperation, probabilistic consensus, and cryptographic anonymization. Meanwhile, cloud computing introduced new constraints and trade-offs for deploying systems as corporations moved on-premise infrastructures to cloud-based virtual machines. It also led to the creation of data center-specific hardware. In this dissertation, I present three distinct technical chapters, each employing a different research style applied to these two contexts. First, I employ a Monte Carlo simulation to recreate and analyze a set of deviant strategies for mining Bitcoin and other popular proof-of-work cryptocurrencies. Next, I present a novel modification to the Linux operating system and the Xen hypervisor that achieves a swap-free ballooning solution for virtual machine migrations when operating in a non-overcommitted cloud machine. Finally, I present unexpected results demonstrating significant performance differences between Amazon Web Services (AWS) cloud-specific graphics cards of the same modeland identify GPU memory as a likely culprit.
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
-
- Negy, Kevin
- Chair dc:contributor.chair
-
- Sampson, Adrian
- Committee members dc:contributor.committeemember
-
- Van Renesse, Robbert
- Pond, Steven
Subjects
dc:subject × 6Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Bibid
- 17422418
- Dc Identifier Other
-
ProQuest Submission ID: 15030
ProQuest Publication ID: 32113619