We herein describe the performance evaluation of a modular implementation of the MGRIT (MultiGrid-In-Time) algorithm within the context of the PETSc (the Portable, Extensible Toolkit for Scientific computing) library. Our aim is to give the PETSc users the opportunity of testing the MGRIT parallel-in-time approach as an alternative to the Time Stepping integrator (TS), when solving their problems arising from the discretization of linear evolutionary models. To this end, we analyzed the performance parameters of the algorithm in order to underline the relationship between the configuration factors and problem characteristics, intentionally overlooking any accuracy issue and spacial parallelism.
Performance evaluation for a PETSc parallel-in-time solver based on the MGRIT algorithm / Mele, V.; Romano, D.; Constantinescu, E. M.; Carracciuolo, L.; D'Amore, L.. - 11339:(2019), pp. 716-728. (Intervento presentato al convegno European Conference on Parallel Processing Euro-Par 2018: Parallel Processing Workshops) [10.1007/978-3-030-10549-5_56].
Performance evaluation for a PETSc parallel-in-time solver based on the MGRIT algorithm
Mele V.
;Romano D.;Carracciuolo L.;D'Amore L.
2019
Abstract
We herein describe the performance evaluation of a modular implementation of the MGRIT (MultiGrid-In-Time) algorithm within the context of the PETSc (the Portable, Extensible Toolkit for Scientific computing) library. Our aim is to give the PETSc users the opportunity of testing the MGRIT parallel-in-time approach as an alternative to the Time Stepping integrator (TS), when solving their problems arising from the discretization of linear evolutionary models. To this end, we analyzed the performance parameters of the algorithm in order to underline the relationship between the configuration factors and problem characteristics, intentionally overlooking any accuracy issue and spacial parallelism.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
