From e11cd9c26d47da75e3ff6ac3e218f644ea6e5314 Mon Sep 17 00:00:00 2001 From: afanfakh Date: Wed, 3 Dec 2014 11:58:41 +0100 Subject: [PATCH] some corrections --- Heter_paper.tex | 17 +++++++++++------ fig/sen_comp.pdf | Bin 17564 -> 17551 bytes 2 files changed, 11 insertions(+), 6 deletions(-) diff --git a/Heter_paper.tex b/Heter_paper.tex index d318912..390b151 100644 --- a/Heter_paper.tex +++ b/Heter_paper.tex @@ -89,7 +89,10 @@ for each node computing the message passing iterative application. The algorithm works without training or profiling. It uses a new energy model for message passing iterative applications running on a heterogeneous platform. The proposed algorithm is evaluated on the Simgrid simulator while running the NAS parallel benchmarks. The experiments demonstrated that it reduces the energy consumption -up to 35\% while limiting the performance degradation as much as possible. +up to 35\% while limiting the performance degradation as much as possible. \textcolor{red}{Furthermore, we compare the +proposed algorithm with other method. The comparison’s results show that our algorithm gives better +energy-time trade-off.} + \end{abstract} \section{Introduction} @@ -127,8 +130,9 @@ the energy-performance objective function that maximizes the reduction of energy consumption while minimizing the degradation of the program's performance. Section~\ref{sec.optim} details the proposed frequency selecting algorithm then the precision of the proposed algorithm is verified. Section~\ref{sec.expe} presents the results of applying the algorithm on the NAS parallel benchmarks and executing them -on a heterogeneous platform. It also shows the results of running three -different power scenarios and comparing them. +on a heterogeneous platform. It shows the results of running three +different power scenarios and comparing them. \textcolor{red}{Moreover, it also shows the comparison results +between our method and other method.} Finally, in Section~\ref{sec.concl} the paper is ended with a summary and some future works. \section{Related works} @@ -1036,7 +1040,7 @@ In this section, we compare our scaling factors selection algorithm with Spiliopoulos et al. algorithm \cite{Spiliopoulos_Green.governors.Adaptive.DVFS}. They developed an online frequency selecting algorithm running over multicore architecture. The algorithm predicted both the energy and performance during the runtime of the program, then -selecting the frequencies that minimized the energy and delay products (EDP), $EDP=Enegry * Delay$. +selecting the frequencies that minimized the energy and delay products (EDP), $EDP=Enegry*Delay$. To be able to compare with this algorithm, we used our energy and execution time models in prediction process, equations (\ref{eq:energy}) and (\ref{eq:fnew}). Also their algorithm is adapted to taking into account the heterogeneous platform to starts selecting the @@ -1077,7 +1081,7 @@ that gave some enhancements to the energy and performance tradeoff. \begin{figure}[t] \centering - \includegraphics[scale=0.6]{fig/compare_EDP.pdf} + \includegraphics[scale=0.5]{fig/compare_EDP.pdf} \caption{Tradeoff comparison for NAS benchmarks class C} \label{fig:compare_EDP} \end{figure} @@ -1088,7 +1092,8 @@ that gave some enhancements to the energy and performance tradeoff. In this paper, a new online frequency selecting algorithm has been presented. It selects the best possible vector of frequency scaling factors that gives the maximum distance (optimal tradeoff) between the predicted energy and the predicted performance curves for a heterogeneous platform. This algorithm uses a new energy model for measuring and predicting the energy of distributed iterative applications running over heterogeneous -platform. To evaluate the proposed method, it was applied on the NAS parallel benchmarks and executed over a heterogeneous platform simulated by Simgrid. The results of the experiments showed that the algorithm reduces up to 35\% the energy consumption of a message passing iterative method while limiting the degradation of the performance. The algorithm also selects different scaling factors according to the percentage of the computing and communication times, and according to the values of the static and dynamic powers of the CPUs. +platform. To evaluate the proposed method, it was applied on the NAS parallel benchmarks and executed over a heterogeneous platform simulated by Simgrid. The results of the experiments showed that the algorithm reduces up to 35\% the energy consumption of a message passing iterative method while limiting the degradation of the performance. The algorithm also selects different scaling factors according to the percentage of the computing and communication times, and according to the values of the static and dynamic powers of the CPUs. \textcolor{red}{ We compare our algorithm with Spiliopoulos et al. algorithm, the comparison results showed that our +algorithm outperforms their algorithm in term of energy-time tradeoff.} In the near future, this method will be applied to real heterogeneous platforms to evaluate its performance in a real study case. It would also be interesting to evaluate its scalability over large scale heterogeneous platform and measure the energy consumption reduction it can produce. Afterward, we would like to develop a similar method that is adapted to asynchronous iterative applications where each task does not wait for others tasks to finish there works. The development of such method might require a new diff --git a/fig/sen_comp.pdf b/fig/sen_comp.pdf index 5d3e407bdc71dbd7742ec6b28dbe3ee26ea32509..f25c19b11f5393c83c4494bc3ddad7ca0b9c9e73 100644 GIT binary patch delta 1423 zcmbQ!$=Kh?IKi`iZ@6#v9Rq>Azr$a|bS%EOJ;6{VbK8XITdi_l`MFVtmQ9HH^xfL{ zQ+vwJn|Ie}s>a&>{`vB3hRLQ>?nhfI#ez;;<9Mu_w6r$+=;`y%UbBB(zEk`>lgPbo zhi=@c+LwFmw!(}1r`NklW-wo#C)Hr~sX)JCMahAM9}o2VKX#cp?~Zx>{@LyPy(-p^ zKC6HG+Erg}aWr}Hnp|1CettVsi*1>E#4~vW_bxV4vUl&C&UwKne%9}sJUT7@w{n<@ zs`-SS4)A=r!V>l{YiX2ifpA3YY@=(%$G^-yKX+5f46f#b&8G{bF1Lj7`pB(6@P#XT zRr;q2)-)Myn_tC|3ldgcV!8HEs=oWFWr5_H!(WB|%kSX1_R!7Pv=3-t&~iy{)yB;` zZ-%X0A;j(eNH=d+)bYj@wcFMTGX%Vu6Y+D0`qH@Br!{Yydk7tIv;8`s#lh9D@X=Ha z!P!wFYgaf5r)={SHYoY${Ihe_0lo_uV-r z@8i?MU3U*}?``4x;W<>Esy((rXRlE;PsC!yBGj+e6+Efzm*6lWmvnDgL+QGHzDr09a>sZ4vn9bD%gqgon!ow3|HV11*IcJ$7 zFW+I}&AQcO9I$5w>skE^TdWTW&73*;5Yx$Xo|h+<>HlM@-QJo$`606mlY!;t|IA4? zOy;JOlO0-s)aI8C!HleiCPqeTSTmU-G BH0}TZ delta 1459 zcmeD0WSrB*-Qa zdlWvzTVD21d*RYB>t9@bXi&r+&n?fjY&Y`qdtZGNPf7tDzJK*`k&sBh z->1gz-`f6{TO18mJ-9LKaRJw2%P^Vz*9L6;?jd=ukC$tFFk+nbGP~4%W@GCI&&j?H z7v~*fS;KGCom~z{wQ{+pX>GV{+q2(4EsTs*Pwp1VO6^R#7PVwfAqU%z5X1jWdzEJ{?X=XEs9&qX zutC>6ca1~iGfVZGiostBG-p5A#hBq?>akeKC5K~Ote4>CqqkIMGHm$Z{4R5g#^j}~ zmb12b?)GBY@qX2mk1S6u`I;jAW+;^H@Z&Uka8GuVu7jNXw>oB~J#*(4cD?pZW!&BP zYR}KW_;n}kU!8sItiB-L^ET(r^&;~`=GCazm+)7ZFdoU?7T@@~#o2C=SjL&j6%9+rqfad>ynO$4`XyhUBDDOC!?)SXl|@YUz6f%ZU9&*MFaM>qVR*$0hsWMa z0{G6gP5n?G`!iLwbcxV|?Tco4&E#ig-E&8LXp8Ykz_J4L-XR%YkO7mZFfjo1g(tMRm%n;p(%mNl{!+um|YQP|1LJu7;FX0<~~l~Z|EH*H;KsV6P)YJ^55JjD-sf8Jan7Ji}UNb`@Ofh3aB+nUvO*S+$HbZx{ zp_!Sf1-d0>=BB2T4?Ek|TROU$8#=nWTDTb)8M<0nH~}daLt_h9XLBv4&D|`3))`nDTDX`wxth3`a@pB& c6_+Fyl~f26rKWKinps*HaH*=g`nz!f01j?0G5`Po -- 2.39.5