4 files

Experimental data and theoretical speedup calculations for Odroid XU3 and Dell XPS platforms

posted on 01.01.2017, 00:00 by A Rafiev, M Al-hayanni, R Shafik, A Romanovsky, A Yakovlev
Traditional speedup models, such as Amdahl's law, have helped the research community and industry better understand system performance capabilities and application parallelizability. As they mostly target homogeneous hardware platforms or limited forms of processor heterogeneity, these models do not cover newly emerging multi-core heterogeneous architectures. In our work, we developed novel speedup and energy consumption models based on a more general representation of heterogeneity, referred to as the normal form heterogeneity, that supports a wide range of heterogeneous many-core architectures. The modelling method aims to predict system power efficiency and performance ranges, and facilitates research and development at the hardware and system software levels. The models were validated through extensive experimentation on the off-the-shelf big.LITTLE heterogeneous platform (Odroid XU3) and a dual-GPU laptop (Dell XPS 15). A quantitative efficiency analysis targeting the system load balancer in Odroid XU3 platform was used to demonstrate the practical use of the method. The presented data contains the raw measurements from these experiments, as well as detailed calculation of the models.