Date of Graduation

5-2026

Document Type

Thesis

Degree Name

Bachelor of Science in Computer Science

Degree Level

Undergraduate

Department

Computer Science and Computer Engineering

Advisor/Mentor

Dr. Alexander H. Nelson

Committee Member

Dr. Alejandro Martin-Gomez

Second Committee Member

Dr. Brajendra Panda

Abstract

Health data is personal, but most apps hand it over to company-owned servers the moment a user hits save. SolidFit challenges that model by storing workout data in a Personal Online Data Store (PDS): a user-controlled repository built on the open Solid protocol. Rather than trusting a platform with their data, users retain direct ownership. This paper evaluates whether a Solid-backed Android application can match the performance of a conventional cloud backend, using Google Firebase as the baseline. Both applications were benchmarked across insert and fetch operations at six payload sizes ranging from 2 KB to 64 KB. Insert latency favors the PDS architecture at larger payloads, with SolidFit completing 64 KB inserts in roughly one third the time Firebase required. Cold cache fetch latencies are comparable across all tested sizes. Warm cache reads favor Firebase, but both implementations remain below human-perceptible thresholds for a fitness tracking use case. Beyond performance, this thesis documents the developer experience gap between the two architectures and shows that annotation-based tooling can automate approximately 73% of the Solid-specific data layer, reducing the integration burden to a manageable level. Together, these results suggest that Solid-based PDS architectures are a viable foundation for privacy-preserving mobile health applications.

Keywords

Solid Protocol; Personal Online Data Stores; Decentralized Mobile Applications; Performance Benchmarking; Android; Health Data Management

Download

Available for download on Friday, October 15, 2027

Share

COinS