Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

BALANCING PRIVACY, PRECISION AND PERFORMANCE IN DISTRIBUTED SYSTEMS

Zaki, Marian (2020) BALANCING PRIVACY, PRECISION AND PERFORMANCE IN DISTRIBUTED SYSTEMS. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

This is the latest version of this item.

[img]
Preview
PDF
Download (5MB) | Preview

Abstract

Privacy, Precision, and Performance (3Ps) are three fundamental design objectives in distributed systems. However, these properties tend to compete with one another and are not considered absolute properties or functions. They must be defined and justified in terms of a system, its resources, stakeholder concerns, and the security threat model.

To date, distributed systems research has only considered the trade-offs of balancing privacy, precision, and performance in a pairwise fashion. However, this dissertation formally explores the space of trade-offs among all 3Ps by examining three representative classes of distributed systems, namely Wireless Sensor Networks (WSNs), cloud systems, and Data Stream Management Systems (DSMSs). These representative systems support large part of the modern and mission-critical distributed systems.

WSNs are real-time systems characterized by unreliable network interconnections and highly constrained computational and power resources. The dissertation proposes a privacy-preserving in-network aggregation protocol for WSNs demonstrating that the 3Ps could be navigated by adopting the appropriate algorithms and cryptographic techniques that are not prohibitively expensive.

Next, the dissertation highlights the privacy and precision issues that arise in cloud databases due to the eventual consistency models of the cloud. To address these issues, consistency enforcement techniques across cloud servers are proposed and the trade-offs between 3Ps are discussed to help guide cloud database users on how to balance these properties.

Lastly, the 3Ps properties are examined in DSMSs which are characterized by high volumes of unbounded input data streams and strict real-time processing constraints. Within this system, the 3Ps are balanced through a proposed simple and efficient technique that applies access control policies over shared operator networks to achieve privacy and precision without sacrificing the systems performance.

Despite that in this dissertation, it was shown that, with the right set of protocols and algorithms, the desirable 3P properties can co-exist in a balanced way in well-established distributed systems, this dissertation is promoting the use of the new 3Ps-by-design concept. This concept is meant to encourage distributed systems designers to proactively consider the interplay among the 3Ps from the initial stages of the systems design lifecycle rather than identifying them as add-on properties to systems.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Zaki, Marianmki6@pitt.edumki6
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairLee, Adamadamlee@cs.pitt.edu
Committee CoChairChrysanthis, Panospanos@cs.pitt.edu
Committee MemberMosse, Danielmosse@pitt.edu
Committee MemberZadorozny, Vladimirviz@pitt.edu
Date: 16 January 2020
Date Type: Publication
Defense Date: 8 November 2019
Approval Date: 16 January 2020
Submission Date: 5 December 2019
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 167
Institution: University of Pittsburgh
Schools and Programs: Dietrich School of Arts and Sciences > Computer Science
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: privacy performance 3Ps-by-design data precision distributed systems
Date Deposited: 16 Jan 2020 20:01
Last Modified: 16 Jan 2020 20:01
URI: http://d-scholarship.pitt.edu/id/eprint/38073

Available Versions of this Item

  • BALANCING PRIVACY, PRECISION AND PERFORMANCE IN DISTRIBUTED SYSTEMS. (deposited 16 Jan 2020 20:01) [Currently Displayed]

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item