Change Safety Net Tool: Improving Live Service Reliability; CrowdStrike and the Risks of Cloud Dependency in Digital Networks

Author:
Spencer, Andrew, School of Engineering and Applied Science, University of Virginia
Advisors:
Elliott, Travis, EN-Engineering and Society, University of Virginia
Vrugtman, Rosanne, EN-Comp Science Dept, University of Virginia
Abstract:

This portfolio presents two independent but complementary investigations into the challenges of building and sustaining reliable digital systems—one focuses on a practical software engineering solution, and the other on broader structural risks in modern digital infrastructures.
My technical project was developed during a summer internship with SWIFT, a global provider of secure financial messaging services. The project addressed the need for greater control and accountability while managing changes on live servers. I helped build a tool—the Change Safety Net (CSN)—that flags potentially disruptive infrastructure changes before they impact live services. Built using Python, PHP, and MySQL, the tool was developed within a scrum framework and deployed to a cross-functional administrative team by the end of my internship. The tool represents a localized, technical solution to an operational problem, and future development may involve machine learning to increase its predictive accuracy and adaptability.
Separately, my STS research investigates the systemic vulnerabilities introduced by reliance on third-party cloud services. With the July 2024 CrowdStrike outage as the central focus, the paper analyzes how a single erroneous update caused global service disruptions due to deep infrastructural interdependence. Through this case, I argue that organizational over-reliance on cloud providers compromises autonomy and resilience, and that technical failures become widespread social issues when mitigation strategies are absent or underdeveloped.
Although the two works address different problems, they both reveal how critical reliability is to digital systems, whether in localized service environments or globally networked infrastructures. One focuses on creating tools to prevent failure; the other on understanding the consequences when such tools or safeguards are absent. Together, they offer insight into how technical decisions and organizational structures each shape the performance and risks of digital networks.

Degree:
BS (Bachelor of Science)
Keywords:
Actor-Network Theory, Cloud Dependency, Reliable digital systems
Notes:

School of Engineering and Applied Science

Bachelor of Science in Computer Science

Technical Advisor: Rosanne Vrugtman

STS Advisor: Travis Elliott

Language:
English
Rights:
All rights reserved (no additional license for public reuse)
Issued Date:
2025/05/09