- Library Home /
- Search Collections /
- Open Collections /
- Browse Collections /
- UBC Theses and Dissertations /
- A middleware framework for quality of service management...
Open Collections
UBC Theses and Dissertations
UBC Theses and Dissertations
A middleware framework for quality of service management and information flow control across the cloud-edge continuum Jung, Kumseok
Abstract
The modern cloud-edge continuum offers a vast, ubiquitous fabric of resources. However, it presents a hostile environment for application developers, characterized by platform heterogeneity, network dynamism, and complex trust boundaries. Existing middleware solutions strive to abstract away these complexities, but often force a trade-off between control and convenience, leading to vendor lock-in in proprietary APIs, ad-hoc configurations, and glue code. This dissertation explores the design space of application frameworks to develop a middleware architecture that strictly decouples infrastructure logic from business logic, focusing on two fundamental, often conflicting, non-functional requirements: Quality of Service (QoS) and Security and Privacy (S&P).
We propose a distributed middleware architecture, the pseudokernel, which intercepts standard POSIX interfaces to support distributed execution without requiring source code modification. By revitalizing the "UNIX philosophy" for the modern landscape, we demonstrate that distributed resource complexity can be encapsulated within existing standard interfaces. To manage QoS, we introduce a transparent layer utilizing an Architecture Description Language (ADL) and an Adaptive Deployment Planner (ADP) to optimize component placement based on high-level goals. To enforce S&P, we develop an efficient Information Flow Control (IFC) layer that combines static analysis with automated code instrumentation, solving the trilemma of transparency, deployability, and efficiency.
Finally, we address the interference between QoS optimization and IFC enforcement through a co-operative integration of the QoS and IFC layers. We demonstrate that conflicting requirements can be resolved through topological restructuring and joint optimization. Our evaluation shows that this architecture achieves strict security compliance with negligible performance overhead, demonstrating that high performance and strong security are not zero-sum trade-offs in cloud-edge middleware.
Item Metadata
| Title |
A middleware framework for quality of service management and information flow control across the cloud-edge continuum
|
| Creator | |
| Supervisor | |
| Publisher |
University of British Columbia
|
| Date Issued |
2026
|
| Description |
The modern cloud-edge continuum offers a vast, ubiquitous fabric of resources. However, it presents a hostile environment for application developers, characterized by platform heterogeneity, network dynamism, and complex trust boundaries. Existing middleware solutions strive to abstract away these complexities, but often force a trade-off between control and convenience, leading to vendor lock-in in proprietary APIs, ad-hoc configurations, and glue code. This dissertation explores the design space of application frameworks to develop a middleware architecture that strictly decouples infrastructure logic from business logic, focusing on two fundamental, often conflicting, non-functional requirements: Quality of Service (QoS) and Security and Privacy (S&P).
We propose a distributed middleware architecture, the pseudokernel, which intercepts standard POSIX interfaces to support distributed execution without requiring source code modification. By revitalizing the "UNIX philosophy" for the modern landscape, we demonstrate that distributed resource complexity can be encapsulated within existing standard interfaces. To manage QoS, we introduce a transparent layer utilizing an Architecture Description Language (ADL) and an Adaptive Deployment Planner (ADP) to optimize component placement based on high-level goals. To enforce S&P, we develop an efficient Information Flow Control (IFC) layer that combines static analysis with automated code instrumentation, solving the trilemma of transparency, deployability, and efficiency.
Finally, we address the interference between QoS optimization and IFC enforcement through a co-operative integration of the QoS and IFC layers. We demonstrate that conflicting requirements can be resolved through topological restructuring and joint optimization. Our evaluation shows that this architecture achieves strict security compliance with negligible performance overhead, demonstrating that high performance and strong security are not zero-sum trade-offs in cloud-edge middleware.
|
| Genre | |
| Type | |
| Language |
eng
|
| Date Available |
2026-04-16
|
| Provider |
Vancouver : University of British Columbia Library
|
| Rights |
Attribution-NonCommercial-NoDerivatives 4.0 International
|
| DOI |
10.14288/1.0451974
|
| URI | |
| Degree (Theses) | |
| Program (Theses) | |
| Affiliation | |
| Degree Grantor |
University of British Columbia
|
| Graduation Date |
2026-05
|
| Campus | |
| Scholarly Level |
Graduate
|
| Rights URI | |
| Aggregated Source Repository |
DSpace
|
Item Media
Item Citations and Data
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International