The Service Oriented Architecture (SOA) paradigm enables the organizations to compose complex business applications from distributed services, supported by third party partners. Usually, Service Level Agreements (SLA) are used to establish the requirements between customers and providers. Efforts to predict the quality of SOA transactions are justified by anticipate possible problems that tends to arise at run-time, disturbing the SLA clauses. In this paper, we proposed a SLA-oriented approach, through stochastic models, to predict resource consumption and service levels degradation in scenarios with different compositions and workloads. Our proposal may be adopted both at design-time and runtime. Measures from a real SOA implementation were used to validate our contributions.
Unfortunately, ACM prohibits us from displaying non-influential references for this paper.
To see the full reference list, please visit http://dl.acm.org/citation.cfm?id=1982518.