Performance characterisation and verification of JavaSpaces based on design of experiments

In the ever increasing world of distributed systems, different middleware implementations can be compared qualitatively or quantitatively. Existing evaluation techniques are often not satisfying. We apply design of experiments (DoE) to evaluate and model the performance of JavaSpaces, a tuple spaces implementation for distributing tasks through a virtual space. DoE is a statistical technique for identifying relevant input factors from a (large) set of input factors. The setup of the experiments is determined using experimental design (ED) and a selection of the relevant input factors is based on analysis of variance (ANOVA). Then regression analysis (RA) is used to obtain a multivariate representation. Some extra experiments are performed to validate this approach.