The data mining field is an important source of large scale applications and datasets which are getting more and more common. In this paper, we present performance evaluations of basic large scale data mining applications on an experimental grid environment. We test the scalability of a new clustering algorithm and a grid-based frequent itemsets generation using a grid workflow. We also compare the performance analysis of the workflow execution to simulated models and give measurements of the overhead related to the workflow engine and the underlying grid environment.