10.17634/123238-3 A Rafiev A Rafiev F Xia F Xia R Gensh R Gensh A Aalsaud A Aalsaud A Romanovsky A Romanovsky A Yakovlev A Yakovlev Models and simulation data supporting the method of Selective Abstraction Newcastle University 2017 Selective abstraction Error contamination Model complexity Stochastic Activity Networks SANs Order Graphs Mobius tool Simulation Odroid XU3 big.LITTLE 2017-01-01 00:00:00 Dataset https://data.ncl.ac.uk/articles/dataset/Models_and_simulation_data_supporting_the_method_of_Selective_Abstraction/10280870 With the increase of system complexity in both platforms and applications, power modelling of heterogeneous systems is facing grand challenges from the model scalability issue. To address these challenges, our work uses two systematic methods: selective abstraction and stochastic techniques. The concept of selective abstraction via black-boxing is realised using hierarchical modelling and cross-layer cuts, respecting the concepts of boxability and error contamination. The stochastic aspect is formally underpinned by Stochastic Activity Networks (SANs). The proposed method is validated with experimental results from Odroid XU3 heterogeneous 8-core platform and is demonstrated to maintain high accuracy while improving scalability.