A multi-core CPU and many-core GPU based fast parallel shuffled complex evolution global optimization approach

Abstract

In the field of hydrological modelling, the global and automatic parameter 1 calibration has been a hot issue for many years. Among automatic parameter 2 optimization algorithms, the shuffled complex evolution developed at the University 3 of Arizona (SCE-UA) is the most successful method for stably and robustly locating 4 the global "best" parameter values. Ever since the invention of the SCE-UA, the 5 profession suddenly has a consistent way to calibrate watershed models. However, the 6 computational efficiency of the SCE-UA significantly deteriorates when coping with 7 big data and complex models. For the purpose of solving the efficiency problem, the 8 recently emerging heterogeneous parallel computing (parallel computing by using the 9 multi-core CPU and many-core GPU) was applied in the parallelization and 10 acceleration of the SCE-UA. The original serial and proposed parallel SCE-UA were 11 compared to test the performance based on the Griewank benchmark function. The 12 comparison results indicated that the parallel SCE-UA converged much faster than the 13 serial version and its optimization accuracy was the same as the serial version. It has a 14 promising application prospect in the field of fast hydrological model parameter 15 optimization. 16

Acceleration of three-dimensional Tokamak magnetohydrodynamical code with graphics processing unit and OpenACC heterogeneous parallel programming

“…The successful application of OpenACC (Kan, Zhang, et al 2016;Kan, Lei, et al 2017;Kan et al 2018;Kan, He, Ding, et al 27 / 31 2017) in CLTx leads to great improvement in computational efficiency for studying the MHD instability in Tokamak. The migration of CLTx from the CPU-MPI platform to GPU-OpenACC platform is relatively easy compared with completely rewriting the code into CUDA or OpenCL.…”

Investigating China’s Urban Air Quality Using Big Data, Information Theory, and Machine Learning

“…The final simulated output is the sum of the estimated output and the estimated output error. The structure of the PEK approximator [31][32][33][34][35][36][37][38][39][40] is shown in Fig. 2, where n1, n2, …, nc denote the number of candidate input variables for Class 1, 2, …, nc, respectively; weight 1, weight 2, …, weight n denote combination weights for component networks; O(s), O(e), and E(e) denote simulated output, estimated output, and estimated output error, respectively.…”

An Improved Coupled Routing and Excess Storage (CREST) Distributed Hydrological Model and Its Verification in Ganjiang River Basin, China

“…The model parameters are optimized using the SCE-UA optimization method developed by Duan [35][36][37][38][39][40][41][42][43][44]. The objective function of the parameter optimization is the Nash-Sutcliffe coefficient of efficiency (NSCE).…”