From: afanfakh Date: Wed, 30 Sep 2015 14:04:46 +0000 (+0200) Subject: adding power senarios section X-Git-Url: https://bilbo.iut-bm.univ-fcomte.fr/and/gitweb/mpi-energy2.git/commitdiff_plain/b4f045bac56831537abd93cb4ee5bc570e53fe3b?hp=-c adding power senarios section --- b4f045bac56831537abd93cb4ee5bc570e53fe3b diff --git a/mpi-energy2-extension/Heter_paper.tex b/mpi-energy2-extension/Heter_paper.tex index d6ff46f..b2c2c21 100644 --- a/mpi-energy2-extension/Heter_paper.tex +++ b/mpi-energy2-extension/Heter_paper.tex @@ -842,6 +842,9 @@ which on average is up 26\%. Therefore, the tradeoff distance is related linear percentage. Finally, the best energy and performance tradeoff depends on the all of the following: 1) the computations to communications ratio when there is a communications and slack times, 2) the differences in computing powers between the computing nodes and 3) the differences in static and the dynamic powers of the nodes.} + + + \subsection{The experimental results of multicores clusters} \label{sec.res-mc} The grid'5000 clusters have different number of cores embedded in their nodes @@ -994,17 +997,107 @@ Scenario name & Cluster name & \begin{tabular}[c]{@{}c@ \label{fig:dist-mc} \end{figure} -\subsection{The results for different power consumption scenarios} -\label{sec.compare} +\subsection{The results of using different static power consumption scenarios} +\label{sec.pow_sen} +The static power consumption for one core of the computing node is the leakage power +consumption when this core is in the idle state. The node's idle state power value that measured +as in section \ref{sec.grid5000} had many power consumptions embedded such as +all cores static powers in addition to the power consumption of the other devices. So, the static power for one core +can't measured precisely. On the other hand, while the static power consumption of +one core representing the core's power when there is no any computation, thus +the majority of ratio of the total power consumption is depends on the dynamic power consumption. +Despite that, the static power consumption is becomes more important when the execution time +increased using DVFS. Therefore, the objective of this section is to verify the ability of the proposed +frequencies selecting algorithm when the static power consumption is changed. + +All the results obtained in the previous sections depend on the measured dynamic power +consumptions as in table \ref{table:grid5000}. Moreover, the static power consumption is assumed for +one core represents 20\% of the measured dynamic power of that core. +This assumption is extended in this section to taking into account others ratios for the static power consumption. +In addition to the previous ratio of the static power consumption, two other scenarios are used which +all of these scenarios can be denoted as follow: +\begin{itemize} +\item 10\% of static power scenario +\item 20\% of static power scenario +\item 30\% of static power scenario +\end{itemize} + +These three scenarios represented the ratio of the static power consumption that can be computed from +the dynamic power consumption of the core. The NAS benchmarks of class D are executed over 16 nodes +in the Nancy site using three clusters: Graphite, Graphene and Griffon. As same as used before, the one site 16 nodes +platform scenario explained in the last experiments, as in table \ref{tab:sc}, is uses to run +the NAS benchmarks with these static power scenarios. + + \begin{figure} + \centering + \includegraphics[scale=0.5]{fig/eng_pow.eps} + \caption{The energy saving percentages for NAS benchmarks of the three power scenario} + \label{fig:eng-pow} +\end{figure} + +\begin{figure} + \centering + \includegraphics[scale=0.5]{fig/per_pow.eps} + \caption{The performance degradation percentages for NAS benchmarks of the three power scenario} + \label{fig:per-pow} +\end{figure} +\begin{figure} + \centering + \includegraphics[scale=0.5]{fig/dist_pow.eps} + \caption{The tradeoff distance for NAS benchmarks of the three power scenario} + \label{fig:dist-pow} +\end{figure} +\begin{figure} + \centering + \includegraphics[scale=0.47]{fig/three_scenarios.pdf} + \caption{Comparing the selected frequencies of MG benchmarks for three static power scenarios} + \label{fig:fre-pow} +\end{figure} -\subsection{The comparison of the proposed scaling algorithm } +The energy saving percentages of NAS benchmarks with these three static power scenarios are presented +in figure \ref{fig:eng_sen}. This figure showed the 10\% of static power scenario +gives the biggest energy saving percentage comparing to 20\% and 30\% static power +scenario. When using smaller ratio of static power consumption, the proposed +frequencies selecting algorithm selects smaller frequencies, bigger scaling factors, +because the static energy consumption not increased significantly the overall energy +consumption. Therefore, more energy reduction can be achieved when the frequencies are scaled down. +For example figure \ref{fig:fre-pow}, illustrated that the proposed algorithm +proportionally scaled down the new computed frequencies with the overall predicted energy +consumption. The results of 30\% static power scenario gives the smallest energy saving percentages +because the new selected frequencies produced smaller ratio in the reduced energy consumption. +Furthermore, The proposed algorithm tries to limit selecting smaller frequencies that increased +the static energy consumption if the static power consumption is increased. +The performance degradation percentages are presented in the figure \ref{fig:per-pow}, +the 30\% of static power scenario had less performance degradation percentage, because +bigger frequencies was selected due to the big ratio in the static power consumption. +The inverse was happens in the 20\% and 30\% scenario, the algorithm was selected +biggest frequencies, smaller scaling factors, according to this increased in the static power ratios. +The tradoff distance for the NAS benchmarks with these three static powers scenarios +are presented in the figure \ref{fig:dist}. The results showed that the tradeoff +distance is the best when the 10\% of static power scenario is used, and this percentage +is decreased for the other two scenarios propositionally to their static power ratios. +In EP benchmarks, the results of energy saving, performance degradation and tradeoff +distance are showed small differences when the these static power scenarios were used, +because this benchmark not has communications. The proposed algorithm is selected +same frequencies in this benchmark when all these static power scenarios are used. +The small differences in the results are due to the static power is consumed during the computation +times side by side to the dynamic power consumption, knowing that the dynamic power consumption +representing the highest ratio in the total power consumption of the core, then any change in +the static power during these times have less affect on the overall energy consumption. While the +inverse was happens for the rest of the benchmarks which have the communications +that increased the static energy consumption linearly to the mount of communications +in these benchmarks. + + + +\subsection{The comparison of the proposed frequencies selecting algorithm } \label{sec.compare_EDP} - + \section{Conclusion} \label{sec.concl} diff --git a/mpi-energy2-extension/fig/dist-eps-converted-to.pdf b/mpi-energy2-extension/fig/dist-eps-converted-to.pdf index ca76725..e0779e3 100644 Binary files a/mpi-energy2-extension/fig/dist-eps-converted-to.pdf and b/mpi-energy2-extension/fig/dist-eps-converted-to.pdf differ diff --git a/mpi-energy2-extension/fig/dist.eps b/mpi-energy2-extension/fig/dist.eps index b903459..b6722e7 100644 --- a/mpi-energy2-extension/fig/dist.eps +++ b/mpi-energy2-extension/fig/dist.eps @@ -1,6 +1,6 @@ %!PS-Adobe-2.0 %%Creator: gnuplot 4.6 patchlevel 6 -%%CreationDate: Fri Sep 11 14:26:55 2015 +%%CreationDate: Wed Sep 30 10:32:15 2015 %%DocumentFonts: (atend) %%BoundingBox: 50 50 554 410 %%Orientation: Portrait @@ -506,7 +506,7 @@ SDict begin [ /Author (afanfakh) % /Producer (gnuplot) % /Keywords () - /CreationDate (Fri Sep 11 14:26:55 2015) + /CreationDate (Wed Sep 30 10:32:15 2015) /DOCINFO pdfmark end } ifelse @@ -872,7 +872,7 @@ LT0 0.10 0.10 0.44 C 1827 3248 327 55 40 45 2 PatternFill 0.10 0.10 0.44 C 892 352 73 1649 40 45 2 PatternFill 0.10 0.10 0.44 C 1438 352 73 1986 40 45 2 PatternFill -0.10 0.10 0.44 C 1984 352 73 1788 40 45 2 PatternFill +0.10 0.10 0.44 C 1984 352 73 1650 40 45 2 PatternFill 0.10 0.10 0.44 C 2530 352 73 1435 40 45 2 PatternFill 0.10 0.10 0.44 C 3076 352 73 2301 40 45 2 PatternFill 0.10 0.10 0.44 C 3622 352 73 2236 40 45 2 PatternFill diff --git a/mpi-energy2-extension/fig/dist_pow.eps b/mpi-energy2-extension/fig/dist_pow.eps index 11fdcc0..cb12b09 100644 --- a/mpi-energy2-extension/fig/dist_pow.eps +++ b/mpi-energy2-extension/fig/dist_pow.eps @@ -1,6 +1,6 @@ %!PS-Adobe-2.0 %%Creator: gnuplot 4.6 patchlevel 6 -%%CreationDate: Fri Sep 11 15:12:32 2015 +%%CreationDate: Tue Sep 29 16:28:50 2015 %%DocumentFonts: (atend) %%BoundingBox: 50 50 554 410 %%Orientation: Portrait @@ -506,7 +506,7 @@ SDict begin [ /Author (afanfakh) % /Producer (gnuplot) % /Keywords () - /CreationDate (Fri Sep 11 15:12:32 2015) + /CreationDate (Tue Sep 29 16:28:50 2015) /DOCINFO pdfmark end } ifelse @@ -846,7 +846,7 @@ LT0 0.10 0.10 0.44 C 1.000 1827 3230 327 55 BoxColFill 0.10 0.10 0.44 C 1.000 921 352 88 2178 BoxColFill 0.10 0.10 0.44 C 1.000 1467 352 88 2060 BoxColFill -0.10 0.10 0.44 C 1.000 2013 352 88 1862 BoxColFill +0.10 0.10 0.44 C 1.000 2013 352 88 1905 BoxColFill 0.10 0.10 0.44 C 1.000 2559 352 88 2335 BoxColFill 0.10 0.10 0.44 C 1.000 3105 352 88 2705 BoxColFill 0.10 0.10 0.44 C 1.000 3651 352 88 2692 BoxColFill @@ -880,7 +880,7 @@ LT0 1.00 0.71 0.76 C 1.000 1827 3010 327 55 BoxColFill 1.00 0.71 0.76 C 1.000 1139 352 88 266 BoxColFill 1.00 0.71 0.76 C 1.000 1685 352 88 1018 BoxColFill -1.00 0.71 0.76 C 1.000 2231 352 88 1832 BoxColFill +1.00 0.71 0.76 C 1.000 2231 352 88 1441 BoxColFill 1.00 0.71 0.76 C 1.000 2777 352 88 359 BoxColFill 1.00 0.71 0.76 C 1.000 3323 352 88 1466 BoxColFill 1.00 0.71 0.76 C 1.000 3869 352 88 1074 BoxColFill diff --git a/mpi-energy2-extension/fig/edp_dist.eps b/mpi-energy2-extension/fig/edp_dist.eps index 39e8637..9d4fd77 100644 --- a/mpi-energy2-extension/fig/edp_dist.eps +++ b/mpi-energy2-extension/fig/edp_dist.eps @@ -1,6 +1,6 @@ %!PS-Adobe-2.0 %%Creator: gnuplot 4.6 patchlevel 6 -%%CreationDate: Fri Sep 11 14:26:16 2015 +%%CreationDate: Wed Sep 30 10:33:44 2015 %%DocumentFonts: (atend) %%BoundingBox: 50 50 554 410 %%Orientation: Portrait @@ -506,7 +506,7 @@ SDict begin [ /Author (afanfakh) % /Producer (gnuplot) % /Keywords () - /CreationDate (Fri Sep 11 14:26:16 2015) + /CreationDate (Wed Sep 30 10:33:44 2015) /DOCINFO pdfmark end } ifelse @@ -915,7 +915,7 @@ LT1 0.10 0.10 0.44 C 1.000 1718 3151 327 55 BoxColFill 0.10 0.10 0.44 C 1.000 983 975 73 1484 BoxColFill 0.10 0.10 0.44 C 1.000 1529 975 73 1787 BoxColFill -0.10 0.10 0.44 C 1.000 2075 975 73 1610 BoxColFill +0.10 0.10 0.44 C 1.000 2075 975 73 1485 BoxColFill 0.10 0.10 0.44 C 1.000 2621 975 73 1292 BoxColFill 0.10 0.10 0.44 C 1.000 3167 975 73 2071 BoxColFill 0.10 0.10 0.44 C 1.000 3713 975 73 2013 BoxColFill diff --git a/mpi-energy2-extension/fig/edp_eng.eps b/mpi-energy2-extension/fig/edp_eng.eps index 45defd5..441318d 100644 --- a/mpi-energy2-extension/fig/edp_eng.eps +++ b/mpi-energy2-extension/fig/edp_eng.eps @@ -1,6 +1,6 @@ %!PS-Adobe-2.0 %%Creator: gnuplot 4.6 patchlevel 6 -%%CreationDate: Fri Sep 11 14:26:03 2015 +%%CreationDate: Wed Sep 30 10:33:54 2015 %%DocumentFonts: (atend) %%BoundingBox: 50 50 554 410 %%Orientation: Portrait @@ -506,7 +506,7 @@ SDict begin [ /Author (afanfakh) % /Producer (gnuplot) % /Keywords () - /CreationDate (Fri Sep 11 14:26:03 2015) + /CreationDate (Wed Sep 30 10:33:54 2015) /DOCINFO pdfmark end } ifelse @@ -871,7 +871,7 @@ LT1 0.10 0.10 0.44 C 1.000 1718 3202 327 55 BoxColFill 0.10 0.10 0.44 C 1.000 983 352 73 2245 BoxColFill 0.10 0.10 0.44 C 1.000 1529 352 73 2540 BoxColFill -0.10 0.10 0.44 C 1.000 2075 352 73 2028 BoxColFill +0.10 0.10 0.44 C 1.000 2075 352 73 1865 BoxColFill 0.10 0.10 0.44 C 1.000 2621 352 73 1796 BoxColFill 0.10 0.10 0.44 C 1.000 3167 352 73 2675 BoxColFill 0.10 0.10 0.44 C 1.000 3713 352 73 2529 BoxColFill diff --git a/mpi-energy2-extension/fig/eng_pow.eps b/mpi-energy2-extension/fig/eng_pow.eps index 24fd7f7..51b0ae1 100644 --- a/mpi-energy2-extension/fig/eng_pow.eps +++ b/mpi-energy2-extension/fig/eng_pow.eps @@ -1,6 +1,6 @@ %!PS-Adobe-2.0 %%Creator: gnuplot 4.6 patchlevel 6 -%%CreationDate: Fri Sep 11 14:25:11 2015 +%%CreationDate: Tue Sep 29 16:30:16 2015 %%DocumentFonts: (atend) %%BoundingBox: 50 50 554 410 %%Orientation: Portrait @@ -506,7 +506,7 @@ SDict begin [ /Author (afanfakh) % /Producer (gnuplot) % /Keywords () - /CreationDate (Fri Sep 11 14:25:11 2015) + /CreationDate (Tue Sep 29 16:30:16 2015) /DOCINFO pdfmark end } ifelse @@ -872,7 +872,7 @@ LT0 0.10 0.10 0.44 C 1.000 1718 3248 327 55 BoxColFill 0.10 0.10 0.44 C 1.000 921 352 88 2602 BoxColFill 0.10 0.10 0.44 C 1.000 1467 352 88 2605 BoxColFill -0.10 0.10 0.44 C 1.000 2013 352 88 1749 BoxColFill +0.10 0.10 0.44 C 1.000 2013 352 88 1788 BoxColFill 0.10 0.10 0.44 C 1.000 2559 352 88 2310 BoxColFill 0.10 0.10 0.44 C 1.000 3105 352 88 2771 BoxColFill 0.10 0.10 0.44 C 1.000 3651 352 88 2706 BoxColFill @@ -889,7 +889,7 @@ LT1 0.24 0.70 0.44 C 1.000 1718 3138 327 55 BoxColFill 0.24 0.70 0.44 C 1.000 1030 352 88 2096 BoxColFill 0.24 0.70 0.44 C 1.000 1576 352 88 2371 BoxColFill -0.24 0.70 0.44 C 1.000 2122 352 88 1753 BoxColFill +0.24 0.70 0.44 C 1.000 2122 352 88 1740 BoxColFill 0.24 0.70 0.44 C 1.000 2668 352 88 1676 BoxColFill 0.24 0.70 0.44 C 1.000 3214 352 88 2497 BoxColFill 0.24 0.70 0.44 C 1.000 3760 352 88 2360 BoxColFill @@ -906,7 +906,7 @@ LT0 1.00 0.71 0.76 C 1.000 1718 3028 327 55 BoxColFill 1.00 0.71 0.76 C 1.000 1139 352 88 671 BoxColFill 1.00 0.71 0.76 C 1.000 1685 352 88 942 BoxColFill -1.00 0.71 0.76 C 1.000 2231 352 88 1745 BoxColFill +1.00 0.71 0.76 C 1.000 2231 352 88 1468 BoxColFill 1.00 0.71 0.76 C 1.000 2777 352 88 326 BoxColFill 1.00 0.71 0.76 C 1.000 3323 352 88 1464 BoxColFill 1.00 0.71 0.76 C 1.000 3869 352 88 1038 BoxColFill diff --git a/mpi-energy2-extension/fig/eng_s-eps-converted-to.pdf b/mpi-energy2-extension/fig/eng_s-eps-converted-to.pdf index 3c4b9ba..c2c7c3e 100644 Binary files a/mpi-energy2-extension/fig/eng_s-eps-converted-to.pdf and b/mpi-energy2-extension/fig/eng_s-eps-converted-to.pdf differ diff --git a/mpi-energy2-extension/fig/eng_s.eps b/mpi-energy2-extension/fig/eng_s.eps index f359b90..0cf21ae 100644 --- a/mpi-energy2-extension/fig/eng_s.eps +++ b/mpi-energy2-extension/fig/eng_s.eps @@ -1,6 +1,6 @@ %!PS-Adobe-2.0 %%Creator: gnuplot 4.6 patchlevel 6 -%%CreationDate: Fri Sep 11 14:27:46 2015 +%%CreationDate: Wed Sep 30 10:37:43 2015 %%DocumentFonts: (atend) %%BoundingBox: 50 50 554 410 %%Orientation: Portrait @@ -506,7 +506,7 @@ SDict begin [ /Author (afanfakh) % /Producer (gnuplot) % /Keywords () - /CreationDate (Fri Sep 11 14:27:46 2015) + /CreationDate (Wed Sep 30 10:37:43 2015) /DOCINFO pdfmark end } ifelse @@ -872,7 +872,7 @@ LT0 0.10 0.10 0.44 C 1827 3248 327 55 40 45 2 PatternFill 0.10 0.10 0.44 C 892 352 73 2096 40 45 2 PatternFill 0.10 0.10 0.44 C 1438 352 73 2371 40 45 2 PatternFill -0.10 0.10 0.44 C 1984 352 73 1893 40 45 2 PatternFill +0.10 0.10 0.44 C 1984 352 73 1740 40 45 2 PatternFill 0.10 0.10 0.44 C 2530 352 73 1676 40 45 2 PatternFill 0.10 0.10 0.44 C 3076 352 73 2497 40 45 2 PatternFill 0.10 0.10 0.44 C 3622 352 73 2360 40 45 2 PatternFill