A Scalable and Highly Available Brokering Service for SLA-Based Composite Services

Abstract

Abstract. The introduction of self-adaptation and self-management techniques in a service-oriented system can allow to meet in a changing environment the levels of service formally defined with the system users in a Service Level Agreement (SLA). However, a self-adaptive SOA system has to be carefully designed in order not to compromise the system scalability and availability. In this paper we present the design and performance evaluation of a brokering service that supports at runtime the self-adaptation of composite services offered to several concurrent users with different service levels. To evaluate the performance of the brokering service, we have carried out an extensive set of experiments on different implementations of the system architecture using workload generators that are based on open and closed system models. The experimental results demonstrate the effectiveness of the brokering service design in achieving scalability and high availability.

A Performance Comparison of QoS-Driven Service Selection Approaches

“…We use Sun BPEL Service Engine for the business process logic, and MATLAB and CPLEX to solve respectively the per-flow and per-request optimization problems. We refer to [4] for a detailed description of the MOSES prototype.…”

“…In the following, we provide an overview of the MOSES system; a detailed description of the per-flow methodology (for whom MOSES has been originally designed) and prototype can be found in [6] and [4], respectively.…”

“…To this end, we consider two representative methodologies that tackle the service selection at the per-request and per-flow grains and incorporate them into the MOSES (MOdel-based SElfadaptation of SOA systems) prototype [4], a runtime adaptation framework for a SOS architected as a service broker. We compare the performance of the two methodologies under two workload scenarios characterized by different workload patterns, considering as main performance metric the fulfillment of the composite service's response time agreed by the broker with its users.…”

Optimal Admission Control for a QoS-Aware Service-Oriented System

“…In this section, we provide an overview on the MOSES system for which we propose in this paper MDP-based admission control policies. A detailed description of the MOSES methodology, architecture and implementing prototype can be found in [7] and [4], respectively. MOSES acts as a third-party intermediary between service users and providers, performing a role of provider towards the users and being in turn a requestor to the providers of the concrete services.…”

“…In this paper, to overcome the limitations of the aforementioned results we study the admission problem for the MOdel-based SElf-adaptation of SOA systems (MOSES) service broker we proposed in [4,7], which manages a composite service offering differentiated QoS service classes. We formulate the optimal admission control problem as a Markov Decision Process (MDP) problem with the goal to maximise the broker discounted expected infinite horizon reward while guaranteeing non-functional QoS requirements to its users.…”