this is for holding javascript data
Xavier Andrade edited Parallelization.tex
over 9 years ago
Commit id: 73e351fe880ffd1c134524ace5f934c545ca3e47
deletions | additions
diff --git a/Parallelization.tex b/Parallelization.tex
index 61606b3..e326f00 100644
--- a/Parallelization.tex
+++ b/Parallelization.tex
...
memory with the objective of using processor caches more efficiently
in finite-difference operators. We have found that by using space
filling curves~\cite{Peano_1890}, as shown in Fig.~\ref{fig:gridmaps},
in particular the Hilbert curve~\cite{Hilbert_1891,Skilling2004} we can produce a significant improvement in the performance of semi-local operations. For example in Fig.\ref{fig:gpu_laplacian} shows that a performance gain of around 50\% can be obtained for the finite-difference Laplacian
operator. operator running on a GPU.
Parallelization in Octopus is performed on different levels. The most
basic one is domain decomposition, were the grid is divided in