high performance computing on graphics processing units: hgpu.org

Artificial neural network (ANN) is widely applied as the data-driven modeling tool in hydroinformatics due to its broad applicability of handling implicit and nonlinear relationships between the input and output data. To obtain a reliable ANN model, training ANN using the data is essential, but the training is usually taking many hours for a large data set and/or for large systems with many variants. This may not be a concern when ANN is trained for offline applications, but it is of great importance when ANN is trained or retrained for real-time and near real-time applications, which are becoming an increasingly interested research theme while the hydroinformatics tools will be an integral part of smart city operation systems. Based on author’s previous research projects, which proved that the GPU-based ANN is over 10 times more efficient than CPU-based ANN for constructing the meta-model to be applied as the surrogate of a physics-based model, this paper presents the latest development of the GPUbased ANN computing kernels that is implemented with OpenCL an Open Compute Language. The generalized ANN can be used as an efficient machine learning library for data-driven modeling. The performance of the implemented library has been tested with the benchmark example of water distribution system modeling and compared with the previous results.