Assessing gains from parallel computation on supercomputers

  1. Lilia Maliar 1
  1. 1 Stanford University
    info

    Stanford University

    Stanford, Estados Unidos

    ROR https://ror.org/00f54p054

Revista:
Working papers = Documentos de trabajo: Serie AD

Ano de publicación: 2013

Número: 10

Páxinas: 1-14

Tipo: Documento de traballo

DIALNET GOOGLE SCHOLAR lock_openAcceso aberto editor

Resumo

We assess gains from parallel computation on Backlight supercomputer. We find that information transfers are expensive. To make parallel computation efficient, a task per core must be sufficiently large, ranging from few seconds to one minute depending on the number of cores employed. For small problems, the shared memory programming (OpenMP) leads to a higher efficiency of parallelization than the distributive memory programming (MPI).

Fonte de datos: Dialnet