Rainfall interception calculation modeling using parallel computation systems on NVIDIA CUDA
Yudi Satria Gondokaryono, Mulkan Fadhli · 2017
The researches on mathematical modeling are getting more complexes and have been applied to various fields, and hydrology is one of the fields that use mathematical modeling on interception. Rainfall interception modeling performed to predict the overflow of water after rain occurs, so it can be used for disaster mitigation in regional development. Calculation of rainfall interception model on land (watershed) which is very spacious and has leafy trees, (eq. garden) required high computational resources (time and computer system). This kind of problems can be overcome by developing the model into parallel computation. In this study, interception modeling on parallel computation performed on a GPU device. CUDA is an Nvidia developed application programming interface (API) that designed to run on the GPU computation. CUDA API allows developed applications to run on CPU and GPU, so it would increase the computation process performance. The results of this study concluded that the performance of CUDA is two times better than the performance of CPU in term of execution time.