Abstract
The software and hardware systems required to deliver modern Web based services are becoming increasingly complex. Management and evolution of the systems requires periodic analysis of performance and capacity to maintain quality of service and maximise efficient use of resources. In this work we present a method that uses a repeated local search technique to improve the accuracy of modelling such systems while also reducing the complexity and time required to perform this task. The accuracy of the model derived from the search-based approach is validated by extrapolating the performance to multiple load levels which enables system capacity and performance to be planned and managed more efficiently. © 2014 Springer International Publishing Switzerland.
Cite
CITATION STYLE
Ghaith, S., Wang, M., Perry, P., & Murphy, J. (2014). Transaction profile estimation of queueing network models for IT systems using a search-based technique. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8636 LNCS, pp. 234–239). Springer Verlag. https://doi.org/10.1007/978-3-319-09940-8_18
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.