Global-driven service composition in mobile and pervasive computing
Bibliographic record
Abstract
Pervasive computing environments enable access to diverse resources and services over networked computing systems. Mobile systems have the potential to be very active participants in such environments as resource providers, since they are in wide-spread use, and can sense and exchange their operating environments' context data. Service-oriented computing has emerged as an important paradigm in pervasive computing, because it packages heterogeneous resources as services that are discoverable, accessible, and reusable. Services offered by potentially multiple devices can be composed to create a new value-added service. Service provision through service composites is explored in this thesis, particularly in pervasive environments where service providers are mobile and communicate with each other in an ad hoc manner. Mobile service providers are free to join and leave a system, making the availability of the services they provide unpredictable. Service execution may fail because of a previously available service provider's absence at runtime. There is significant potential for improving overall service quality in real-time services provisioning by re-composing better services from the environment including those that may have appeared even during service execution. Mobility also changes the network topology and the links between services, which can lead to execution path loss, and in turn composition failures at runtime. Thus, service composition requires a comprehensive and dynamic discovery model to reason about an appropriate combination of services that match the given functionality, as well as an efficient mechanism that adapts composite services to dynamic environments. Existing research on service-oriented computing has led to automatic planning, adaptive composition and composition recovery to tackle dynamic environments, but requires global service knowledge or a view of the real-time service links. Given mobile devices' limited communication ranges, the network topology changes quickly when devices are roaming, and keeping such system views up-to-date leads to additional communication, maintenance overhead, and may delay the composition process. This thesis presents a fully decentralized services composition model that supports flexible service discovery and execution in mobile pervasive environments. The model is goal-driven, focusing on time-efficient service provisioning to reduce the interference of topology changes. This goal-driven approach achieves flexible service discovery by dynamically planning a service workflow, which supports not only sequential service composites but also complex composites such as parallel or hybrid service flows. Service links' reliability and quality of service issues are considered when selecting services for invocation, which reduces the possibility of execution failures and the effort required for maintaining backup services for composition recovery. If necessary, failure recovery is attempted by adaptable OR-split transitions in the service workflow. The model has been evaluated using both simulation and a prototype case study. Evaluation metrics include measurements of composition success rates under various mobility models, and the composition model's scalability and performance. Simulation results illustrate both the strengths and the limitations of the proposed mechanism in dynamic pervasive computing environments, under different network density and composite complexity conditions. The prototype case study demonstrates this approach's feasibility on real mobile devices.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".