This work presents an extension of the popular EPANET toolkit that aims to achieve shorter execution time of hydraulic simulations by utilising the high performance of graphics processing units (GPUs). This extension implements a Conjugate Gradient Method that performs the computation on a GPU using sparse matrices. The experimental results using real-life networks show that there is considerable potential to reduce the time for repetitive running of the EPANET models by using the GPU technology.
High performance hydraulic simulations with EPANET on Graphics Processing Units / Guidolin, M; Savic, Da; Kapelan, Z; Giustolisi, Orazio. - 2:(2010), pp. 897-904. (Intervento presentato al convegno 9th International Conference on Hydroinformatics tenutosi a Tianjin (Cina) nel 7-10 September 2010).
High performance hydraulic simulations with EPANET on Graphics Processing Units
GIUSTOLISI, Orazio
2010-01-01
Abstract
This work presents an extension of the popular EPANET toolkit that aims to achieve shorter execution time of hydraulic simulations by utilising the high performance of graphics processing units (GPUs). This extension implements a Conjugate Gradient Method that performs the computation on a GPU using sparse matrices. The experimental results using real-life networks show that there is considerable potential to reduce the time for repetitive running of the EPANET models by using the GPU technology.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
