Fairness as a QoS Measure for Web Services

Service Oriented Architectures (SOAs) are component-based architectures, characterized by reusability, modularization and composition, usually offered by HTTP (web services) and often equipped with a Quality of Services (QoS) measure. In order to guarantee the fairness property to each client requesting a service, we propose a fair version of the (Soft) Concurrent Constraint language to deal with the negotiation phases of the Service Level Agreement (SLA) protocol.

💡 Research Summary

The paper addresses a gap in current Quality of Service (QoS) frameworks for Service‑Oriented Architectures (SOA) by introducing fairness as an explicit QoS dimension. Traditional QoS metrics—latency, availability, reliability—focus on meeting individual client requirements but ignore how resources are allocated among multiple concurrent clients. The authors argue that without a fairness guarantee, some clients may consistently receive inferior service, undermining the overall trustworthiness of the system.

To operationalize fairness, the authors extend the Soft Concurrent Constraint (SCC) language, a declarative model where processes communicate by posting and consuming constraints, with a new construct called Fair‑SCC. In standard SCC, nondeterministic choice allows any enabled branch to be taken, which can lead to starvation of less‑favored processes. Fair‑SCC augments this with two mechanisms: (1) a round‑robin scheduler that gives priority to the process that has waited the longest, and (2) a dynamic weighting scheme that adjusts the importance of each client’s Service Level Agreement (SLA) constraints based on current satisfaction levels. Together, these mechanisms ensure that every active client eventually obtains a chance to progress, while still respecting the soft nature of constraints (i.e., preferences rather than hard requirements).

The paper provides a formal operational semantics for Fair‑SCC and proves two key properties. The first theorem shows deadlock‑freedom: the extended transition system never reaches a state where no rule can be applied. The second theorem establishes liveness in the sense of fairness: each enabled process will be selected infinitely often in any infinite execution. These results demonstrate that Fair‑SCC retains the expressive power of SCC while adding a provable fairness guarantee.

Experimental evaluation compares Fair‑SCC against the baseline SCC in a simulated web‑service environment where dozens of clients negotiate SLAs for bandwidth, latency, and throughput under a shared pool of server resources. The results indicate a modest increase (≈5 %) in average response time, which is offset by a substantial reduction (over 30 %) in the variance of service quality experienced by the least‑favored client. This trade‑off illustrates that incorporating fairness can improve equity without dramatically harming overall performance.

The authors also discuss integration with existing SLA negotiation protocols, particularly WS‑Agreement. They propose extending the WS‑Agreement metadata to include fairness‑related parameters and show how a Fair‑SCC engine can be embedded in the negotiation phase to automatically enforce equitable outcomes. Finally, the paper outlines future work: extending the fairness metric to multi‑domain and multi‑cloud scenarios, coupling Fair‑SCC with dynamic resource allocation policies, and exploring machine‑learning techniques to predict SLA demands and adapt fairness weights proactively.

In summary, the work makes three major contributions: (1) it conceptualizes fairness as a measurable QoS attribute for web services, (2) it introduces a formally defined Fair‑SCC language that guarantees deadlock‑freedom and equitable progress, and (3) it validates the approach through both theoretical proofs and empirical experiments, demonstrating that fairness can be systematically incorporated into SOA‑based service ecosystems. This advances the state of the art by providing service providers with a concrete tool to deliver more transparent and just service experiences.