Dynamic QoS-Aware Coalition Formation

Abstract

Users of wireless devices increasingly demand access to multimedia content with specific quality of service requirements. Users might tolerate different levels of service, or could be satisfied with different quality combinations choices. However, multimedia processing introduces heavy resource requirements on the client side.Our work tries to address the growing demand on resources and performance requirements, by allowing wireless nodes to cooperate with each other to meet resource allocation requests and handle stringent constraints, opportunistically taking advantage of the local ad-hoc network that is created spontaneously, as nodes move in range of each other, forming a temporary coalition for service execution. Coalition formation is necessary when a single node cannot execute a specific service, but it may also be beneficial when groups perform more efficiently when compared to a single s node performance.

Time-bounded distributed QoS-aware service configuration in heterogeneous cooperative environments

“…The main goal of the conducted evaluations was to measure the improvement in the performance of the CooperatES framework when operating in open and dynamic environments with respect to its previous version, where a traditional QoS optimisation approach was used [7]. In Section 8.2, the performance profiles of the proposed anytime algorithms are determined and Section 8.3 discusses the computational cost of the proposed anytime algorithms when compared against our previous traditional QoS optimisation approach.…”

“…While we certainly share some concerns with these works, and also apply utility-based adaptation strategies [7], we go a step further and propose a QoS-aware cooperative service execution to deal with a large number of tasks, multiple resources, and highly dynamic real-time operation constraints in open real-time systems.…”

“…After evaluating all candidate proposals the anytime version of the algorithm will produce exactly the same solution quality as its traditional version [7] that only produces a solution with quality Q coalition at the end of computation. If the time required to evaluate a candidate proposal is t e , the total required runtime of the anytime algorithm is the sum of all n evaluations.…”

“…When it runs to completion, the anytime version of the algorithm will produce exactly the same solution as its traditional version [7] that only produces a solution with quality Q c onf at the end of its computation time.…”

“…However, it is important to analyse the computational cost required by this approach to reach its optimal solution when compared against the traditional versions of those algorithms [7].…”