Sagas calculi have been proposed to specify distributed Long Running Transactions (LRT) and, in previous work, a subset of naive sagas has been encoded in the Signal Calculus (SC) to enable their use in service-oriented systems. Here, we promote a formal approach to the refactoring of LRT represented in SC so that distributed LRT designed in the Business Process Modelling Notation (BPMN) can be faithfully represented. Firstly, we complete the initial encoding of naive sagas into SC. Secondly, on top of SC, we define a few refactoring transformations for distributed LRT. Finally, we prove that the given refactoring rules are sound by showing that they preserve (weak) bisimilarity.
CITATION STYLE
Ferrari, G. L., Guanciale, R., Strollo, D., & Tuosto, E. (2009). Refactoring long running transactions. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5387, pp. 207–223). https://doi.org/10.1007/978-3-642-01364-5_13
Mendeley helps you to discover research relevant for your work.