Skip to Main Content
Enterprise Service Oriented Architecture (ESOA) is taking centre stage in the transformation journey of large corporations due to its extravagant promise of dynamic composition of applications from services. As the penetration deepens, an interesting pattern emerges: SOA becomes the new silver bullet; initiatives are pushed to run under the SOA banner, and more and more fail to deliver the promise especially on performance. The authors propose an Evidence-informed Layered Queue Model (EI-LQM), abstracting from real world failure and success. It serves to optimize performance for ESOA applications by measuring, tuning and predicting utilizing a formula that parameterizes the patterns of a replication process in large scale ESOA initiatives, where performance has been one of the chronic aspects that is recalcitrant but must be continuously managed. This paper starts the conceptualization of a new model for performance management of ESOA.
Date of Conference: 4-6 Jan. 2013