Merge branch 'master' of ssh://info.iut-bm.univ-fcomte.fr/mpi-energy

[mpi-energy.git] / paper.tex

diff --git a/paper.tex b/paper.tex
index 48f38b1bb055511dc87c9569825dc980d0800d6d..53a1cf78d501428bf64ca4d5bad550671ceba416 100644 (file)
--- a/paper.tex
+++ b/paper.tex
@@ -181,7 +181,7 @@ using a multimeter, the slack times, \dots{} Then a method will exploit these
 measurements to compute the scaling factor values for each processor.  This
 operation, measurements and computing new scaling factor, can be repeated as
 much as needed if the iterations are not regular.  Kimura, Peraza, Yu-Liang et
 measurements to compute the scaling factor values for each processor.  This
 operation, measurements and computing new scaling factor, can be repeated as
 much as needed if the iterations are not regular.  Kimura, Peraza, Yu-Liang et

-al.~\cite{11,2,31} used learning methods to select the appropriate scaling
+al.~\cite{11,2,31} used varied heuristics to select the appropriate scaling

 factor values to eliminate the slack times during runtime.  However, as seen
 in~\cite{39,19}, machine learning methods can take a lot of time to converge
 when the number of available gears is big.  To reduce the impact of slack times,
 factor values to eliminate the slack times during runtime.  However, as seen
 in~\cite{39,19}, machine learning methods can take a lot of time to converge
 when the number of available gears is big.  To reduce the impact of slack times,