Omo: Optimize mapreduce overlap with a good start (reduce) and a good finish (map)

J Wang, Y Yao, Y Mao, B Sheng… - 2015 IEEE 34th …, 2015 - ieeexplore.ieee.org
MapReduce has become a popular data processing framework in the past few years.
Scheduling algorithm is crucial to the performance of a MapReduce cluster, especially when …

I/O performance of the Santos Dumont supercomputer

JL Bez, AR Carneiro, PJ Pavan… - … Journal of High …, 2020 - journals.sagepub.com
In this article, we study the I/O performance of the Santos Dumont supercomputer, since the
gap between processing and data access speeds causes many applications to spend a …

A Hardware/Software Codesign for the Chemical Reactivity of BRAMS

CAO De Souza, EDS Pereira… - … Conference on Digital …, 2017 - ieeexplore.ieee.org
BRAMS is a legacy high-performance computing application responsible for weather
forecasting in Brazil. Where 90% of its code is in Fortran 90, and the remaining in C. In this …

Exploiting multiple levels of parallelism and online refinement of unstructured meshes in atmospheric model application

C Schepke - 2012 - lume.ufrgs.br
Weather forecasts for long periods of time has emerged as increasingly important. The
global concern with the consequences of climate changes has stimulated researches to …

Estudo exploratório sobre fontes de os jitter no kernel linux

EV Paulo Sobrinho - 2012 - repositorio.ufu.br
O fenômeno OS Jitter tem sido considerado um importante fator na computação distribuída
de alto desempenho. Neste trabalho, apresentam-se os resultados de um estudo …

Estudo exploratório sobre fontes de os jitter no kernel linux

P Sobrinho - 2012 - bdtd.ibict.br
The OS Jitter phenomenon has been considered an important factor in high-performance
distributed computing. In this thesis, we present the results of an experimental study that …