From: afanfakh Date: Mon, 1 Feb 2016 11:30:34 +0000 (+0100) Subject: adding some missing files X-Git-Url: https://bilbo.iut-bm.univ-fcomte.fr/and/gitweb/mpi-energy2.git/commitdiff_plain/58911cabccb65d874a5a354483718b07fd0aba3d?ds=inline adding some missing files --- diff --git a/Heter_paper.tex b/Heter_paper.tex index 2f78db5..1ac7dcf 100644 --- a/Heter_paper.tex +++ b/Heter_paper.tex @@ -53,8 +53,6 @@ \newcommand{\Sopt}[1][]{\Xsub{S}{opt}_{#1}} \newcommand{\Tcm}[1][]{\Xsub{T}{cm}_{\fxheight{#1}}} \newcommand{\Tcp}[1][]{\Xsub{T}{cp}_{#1}} -\newcommand{\Ppeak}[1][]{\Xsub{P}{peak}_{#1}} -\newcommand{\Pidle}[1][]{\Xsub{P}{idle}_{\fxheight{#1}}} \newcommand{\TcpOld}[1][]{\Xsub{T}{cpOld}_{#1}} \newcommand{\Tnew}{\Xsub{T}{New}} \newcommand{\Told}{\Xsub{T}{Old}} @@ -85,18 +83,168 @@ \maketitle \begin{abstract} - + Computing platforms are consuming more and more energy due to the increasing + number of nodes composing them. To minimize the operating costs of these + platforms many techniques have been used. Dynamic voltage and frequency + scaling (DVFS) is one of them. It reduces the frequency of a CPU to lower its + energy consumption. However, lowering the frequency of a CPU may increase + the execution time of an application running on that processor. Therefore, + the frequency that gives the best trade-off between the energy consumption and + the performance of an application must be selected. + + In this paper, a new online frequency selecting algorithm for heterogeneous + platforms (heterogeneous CPUs) is presented. It selects the frequencies and tries to give the best + trade-off between energy saving and performance degradation, for each node + computing the message passing iterative application. The algorithm has a small + overhead and 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 show that it reduces the + energy consumption by up to \np[\%]{34} while limiting the performance + degradation as much as possible. Finally, the algorithm is compared to an + existing method, the comparison results show that it outperforms the + latter, on average it saves \np[\%]{4} more energy while keeping the same performance. \end{abstract} \section{Introduction} \label{sec.intro} - +The need for more computing power is continually increasing. To partially +satisfy this need, most supercomputers constructors just put more computing +nodes in their platform. The resulting platforms may achieve higher floating +point operations per second (FLOPS), but the energy consumption and the heat +dissipation are also increased. As an example, the Chinese supercomputer +Tianhe-2 had the highest FLOPS in November 2014 according to the Top500 list +\cite{TOP500_Supercomputers_Sites}. However, it was also the most power hungry +platform with its over 3 million cores consuming around 17.8 megawatts. +Moreover, according to the U.S. annual energy outlook 2014 +\cite{U.S_Annual.Energy.Outlook.2014}, the price of energy for 1 megawatt-hour +was approximately equal to \$70. Therefore, the price of the energy consumed by +the Tianhe-2 platform is approximately more than \$10 million each year. The +computing platforms must be more energy efficient and offer the highest number +of FLOPS per watt possible, such as the L-CSC from the GSI Helmholtz Center +which became the top of the Green500 list in November 2014 \cite{Green500_List}. +This heterogeneous platform executes more than 5 GFLOPS per watt while consuming +57.15 kilowatts. + +Besides platform improvements, there are many software and hardware techniques +to lower the energy consumption of these platforms, such as scheduling, DVFS, +\dots{} DVFS is a widely used process to reduce the energy consumption of a +processor by lowering its frequency +\cite{Rizvandi_Some.Observations.on.Optimal.Frequency}. However, it also reduces +the number of FLOPS executed by the processor which may increase the execution +time of the application running over that processor. Therefore, researchers use +different optimization strategies to select the frequency that gives the best +trade-off between the energy reduction and performance degradation ratio. In +\cite{Our_first_paper}, a frequency selecting algorithm was proposed to reduce +the energy consumption of message passing iterative applications running over +homogeneous platforms. The results of the experiments show significant energy +consumption reductions. In this paper, a new frequency selecting algorithm +adapted for heterogeneous platform is presented. It selects the vector of +frequencies, for a heterogeneous platform running a message passing iterative +application, that simultaneously tries to offer the maximum energy reduction and +minimum performance degradation ratio. The algorithm has a very small overhead, +works online and does not need any training or profiling. + +This paper is organized as follows: Section~\ref{sec.relwork} presents some +related works from other authors. Section~\ref{sec.exe} describes how the +execution time of message passing programs can be predicted. It also presents +an energy model that predicts the energy consumption of an application running +over a heterogeneous platform. Section~\ref{sec.compet} presents 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 shows the results of running +three different power scenarios and comparing them. Moreover, it also shows the +comparison results between the proposed method and an existing method. Finally, +in Section~\ref{sec.concl} the paper ends with a summary and some future works. \section{Related works} \label{sec.relwork} +DVFS is a technique used in modern processors to scale down both the voltage and +the frequency of the CPU while computing, in order to reduce the energy +consumption of the processor. DVFS is also allowed in GPUs to achieve the same +goal. Reducing the frequency of a processor lowers its number of FLOPS and may +degrade the performance of the application running on that processor, especially +if it is compute bound. Therefore selecting the appropriate frequency for a +processor to satisfy some objectives, while taking into account all the +constraints, is not a trivial operation. Many researchers used different +strategies to tackle this problem. Some of them developed online methods that +compute the new frequency while executing the application, such +as~\cite{Hao_Learning.based.DVFS,Spiliopoulos_Green.governors.Adaptive.DVFS}. +Others used offline methods that may need to run the application and profile +it before selecting the new frequency, such +as~\cite{Rountree_Bounding.energy.consumption.in.MPI,Cochran_Pack_and_Cap_Adaptive_DVFS}. +The methods could be heuristics, exact or brute force methods that satisfy +varied objectives such as energy reduction or performance. They also could be +adapted to the execution's environment and the type of the application such as +sequential, parallel or distributed architecture, homogeneous or heterogeneous +platform, synchronous or asynchronous application, \dots{} + +In this paper, we are interested in reducing energy for message passing +iterative synchronous applications running over heterogeneous platforms. Some +works have already been done for such platforms and they can be classified into +two types of heterogeneous platforms: +\begin{itemize} +\item the platform is composed of homogeneous GPUs and homogeneous CPUs. +\item the platform is only composed of heterogeneous CPUs. +\end{itemize} + +For the first type of platform, the computing intensive parallel tasks are +executed on the GPUs and the rest are executed on the CPUs. Luley et +al.~\cite{Luley_Energy.efficiency.evaluation.and.benchmarking}, proposed a +heterogeneous cluster composed of Intel Xeon CPUs and NVIDIA GPUs. Their main +goal was to maximize the energy efficiency of the platform during computation by +maximizing the number of FLOPS per watt generated. +In~\cite{KaiMa_Holistic.Approach.to.Energy.Efficiency.in.GPU-CPU}, Kai Ma et +al. developed a scheduling algorithm that distributes workloads proportional to +the computing power of the nodes which could be a GPU or a CPU. All the tasks +must be completed at the same time. In~\cite{Rong_Effects.of.DVFS.on.K20.GPU}, +Rong et al. showed that a heterogeneous (GPUs and CPUs) cluster that enables +DVFS gave better energy and performance efficiency than other clusters only +composed of CPUs. + +The work presented in this paper concerns the second type of platform, with +heterogeneous CPUs. Many methods were conceived to reduce the energy +consumption of this type of platform. Naveen et +al.~\cite{Naveen_Power.Efficient.Resource.Scaling} developed a method that +minimizes the value of $\mathit{energy}\times \mathit{delay}^2$ (the delay is +the sum of slack times that happen during synchronous communications) by +dynamically assigning new frequencies to the CPUs of the heterogeneous cluster. +Lizhe et al.~\cite{Lizhe_Energy.aware.parallel.task.scheduling} proposed an +algorithm that divides the executed tasks into two types: the critical and non +critical tasks. The algorithm scales down the frequency of non critical tasks +proportionally to their slack and communication times while limiting the +performance degradation percentage to less than \np[\%]{10}. +In~\cite{Joshi_Blackbox.prediction.of.impact.of.DVFS}, they developed a +heterogeneous cluster composed of two types of Intel and AMD processors. They +use a gradient method to predict the impact of DVFS operations on performance. +In~\cite{Shelepov_Scheduling.on.Heterogeneous.Multicore} and +\cite{Li_Minimizing.Energy.Consumption.for.Frame.Based.Tasks}, the best +frequencies for a specified heterogeneous cluster are selected offline using +some heuristic. Chen et +al.~\cite{Chen_DVFS.under.quality.of.service.requirements} used a greedy dynamic +programming approach to minimize the power consumption of heterogeneous servers +while respecting given time constraints. This approach had considerable +overhead. In contrast to the above described papers, this paper presents the +following contributions : +\begin{enumerate} +\item two new energy and performance models for message passing iterative + synchronous applications running over a heterogeneous platform. Both models + take into account communication and slack times. The models can predict the + required energy and the execution time of the application. + +\item a new online frequency selecting algorithm for heterogeneous + platforms. The algorithm has a very small overhead and does not need any + training or profiling. It uses a new optimization function which + simultaneously maximizes the performance and minimizes the energy consumption + of a message passing iterative synchronous application. + +\end{enumerate} \section{The performance and energy consumption measurements on heterogeneous architecture} \label{sec.exe} @@ -106,9 +254,11 @@ In this paper, we are interested in reducing the energy consumption of message passing distributed iterative synchronous applications running over -heterogeneous grid platforms. A heterogeneous grid platform could be defined as a collection of -heterogeneous computing clusters interconnected via a long distance network which has lower bandwidth -and higher latency than the local networks of the clusters. Each computing cluster in the grid is composed of homogeneous nodes that are connected together via high speed network. Therefore, each cluster has different characteristics such as computing power (FLOPS), energy consumption, CPU's frequency range, network bandwidth and latency. +heterogeneous platforms. A heterogeneous platform is defined as a collection of +heterogeneous computing nodes interconnected via a high speed homogeneous +network. Therefore, each node has different characteristics such as computing +power (FLOPS), energy consumption, CPU's frequency range, \dots{} but they all +have the same network bandwidth and latency. \begin{figure}[!t] \centering @@ -117,13 +267,13 @@ and higher latency than the local networks of the clusters. Each computing clust \label{fig:heter} \end{figure} -The overall execution time of a distributed iterative synchronous application -over a heterogeneous grid consists of the sum of the computation time and +The overall execution time of a distributed iterative synchronous application +over a heterogeneous platform consists of the sum of the computation time and the communication time for every iteration on a node. However, due to the -heterogeneous computation power of the computing clusters, slack times may occur +heterogeneous computation power of the computing nodes, slack times may occur when fast nodes have to wait, during synchronous communications, for the slower nodes to finish their computations (see Figure~\ref{fig:heter}). Therefore, the -overall execution time of the program is the execution time of the slowest task +overall execution time of the program is the execution time of the slowest task which has the highest computation time and no slack time. Dynamic Voltage and Frequency Scaling (DVFS) is a process, implemented in @@ -149,39 +299,40 @@ communication time for a task is the summation of periods of time that begin with an MPI call for sending or receiving a message until the message is synchronously sent or received. -Since in a heterogeneous grid each cluster has different characteristics, -especially different frequency gears, when applying DVFS operations on the nodes -of these clusters, they may get different scaling factors represented by a scaling vector: -$(S_{11}, S_{12},\dots, S_{NM})$ where $S_{ij}$ is the scaling factor of processor $j$ in cluster $i$ . To +Since in a heterogeneous platform each node has different characteristics, +especially different frequency gears, when applying DVFS operations on these +nodes, they may get different scaling factors represented by a scaling vector: +$(S_1, S_2,\dots, S_N)$ where $S_i$ is the scaling factor of processor $i$. To be able to predict the execution time of message passing synchronous iterative -applications running over a heterogeneous grid, for different vectors of +applications running over a heterogeneous platform, for different vectors of scaling factors, the communication time and the computation time for all the tasks must be measured during the first iteration before applying any DVFS operation. Then the execution time for one iteration of the application with any vector of scaling factors can be predicted using (\ref{eq:perf}). \begin{equation} \label{eq:perf} - \Tnew = \mathop{\max_{i=1,\dots N}}_{j=1,\dots,M}({\TcpOld[ij]} \cdot S_{ij}) - +\mathop{\min_{j=1,\dots,M}} (\Tcm[hj]) + \Tnew = \max_{i=1,2,\dots,N} ({\TcpOld[i]} \cdot S_{i}) + \MinTcm \end{equation} - -where $N$ is the number of clusters in the grid, $M$ is the number of nodes in -each cluster, $\TcpOld[ij]$ is the computation time of processor $j$ in the cluster $i$ -and $\Tcm[hj]$ is the communication time of processor $j$ in the cluster $h$ during the -first iteration. The model computes the maximum computation time with scaling factor -from each node added to the communication time of the slowest node in the slowest cluster $h$. -It means only the communication time without any slack time is taken into account. -Therefore, the execution time of the iterative application is equal to +Where: +\begin{equation} + \label{eq:perf2} + \MinTcm = \min_{i=1,2,\dots,N} (\Tcm[i]) +\end{equation} +where $\TcpOld[i]$ is the computation time of processor $i$ during the first +iteration and $\MinTcm$ is the communication time of the slowest processor from +the first iteration. The model computes the maximum computation time with +scaling factor from each node added to the communication time of the slowest +node. It means only the communication time without any slack time is taken into +account. Therefore, the execution time of the iterative application is equal to the execution time of one iteration as in (\ref{eq:perf}) multiplied by the number of iterations of that application. This prediction model is developed from the model to predict the execution time -of message passing distributed applications for homogeneous and heterogeneous clusters -~\cite{Our_first_paper,pdsec2015}. The execution time prediction model is +of message passing distributed applications for homogeneous +architectures~\cite{Our_first_paper}. The execution time prediction model is used in the method to optimize both the energy consumption and the performance of iterative methods, which is presented in the following sections. - \subsection{Energy model for heterogeneous platform} Many researchers~\cite{Malkowski_energy.efficient.high.performance.computing, @@ -264,32 +415,31 @@ processor after scaling its frequency is computed as follows: \Es = \Ps \cdot (\Tcp \cdot S + \Tcm) \end{equation} -In the considered heterogeneous grid platform, each node $j$ in cluster $i$ may have -different dynamic and static powers from the nodes of the other clusters, -noted as $\Pd[ij]$ and $\Ps[ij]$ respectively. Therefore, even if the distributed -message passing iterative application is load balanced, the computation time of each CPU $j$ -in cluster $i$ noted $\Tcp[ij]$ may be different and different frequency scaling factors may be +In the considered heterogeneous platform, each processor $i$ may have +different dynamic and static powers, noted as $\Pd[i]$ and $\Ps[i]$ +respectively. Therefore, even if the distributed message passing iterative +application is load balanced, the computation time of each CPU $i$ noted +$\Tcp[i]$ may be different and different frequency scaling factors may be computed in order to decrease the overall energy consumption of the application -and reduce slack times. The communication time of a processor $j$ in cluster $i$ is noted as -$\Tcm[ij]$ and could contain slack times when communicating with slower nodes, +and reduce slack times. The communication time of a processor $i$ is noted as +$\Tcm[i]$ and could contain slack times when communicating with slower nodes, see Figure~\ref{fig:heter}. Therefore, all nodes do not have equal communication times. While the dynamic energy is computed according to the frequency scaling factor and the dynamic power of each node as in (\ref{eq:Edyn}), the static energy is computed as the sum of the execution time of one iteration multiplied by the static power of each processor. The overall energy consumption of a message passing distributed application executed over a -heterogeneous grid platform during one iteration is the summation of all dynamic and -static energies for $M$ processors in $N$ clusters. It is computed as follows: +heterogeneous platform during one iteration is the summation of all dynamic and +static energies for each processor. It is computed as follows: \begin{multline} \label{eq:energy} - E = \sum_{i=1}^{N} \sum_{i=1}^{M} {(S_{ij}^{-2} \cdot \Pd[ij] \cdot \Tcp[ij])} + - \sum_{i=1}^{N} \sum_{j=1}^{M} (\Ps[ij] \cdot {} \\ - (\mathop{\max_{i=1,\dots N}}_{j=1,\dots,M}({\Tcp[ij]} \cdot S_{ij}) - +\mathop{\min_{j=1,\dots M}} (\Tcm[hj]) )) + E = \sum_{i=1}^{N} {(S_i^{-2} \cdot \Pd[i] \cdot \Tcp[i])} + {} \\ + \sum_{i=1}^{N} (\Ps[i] \cdot (\max_{i=1,2,\dots,N} (\Tcp[i] \cdot S_{i}) + + {\MinTcm))} \end{multline} Reducing the frequencies of the processors according to the vector of scaling -factors $(S_{11}, S_{12},\dots, S_{NM})$ may degrade the performance of the application +factors $(S_1, S_2,\dots, S_N)$ may degrade the performance of the application and thus, increase the static energy because the execution time is increased~\cite{Kim_Leakage.Current.Moore.Law}. The overall energy consumption for the iterative application can be measured by measuring the energy @@ -312,11 +462,11 @@ characteristics of each processor (computation power, range of frequencies, dynamic and static powers) and the task executed (computation/communication ratio). The aim being to reduce the overall energy consumption and to avoid increasing significantly the execution time. In our previous -work~\cite{Our_first_paper,pdsec2015}, we proposed a method that selects the optimal -frequency scaling factor for a homogeneous and heterogeneous clusters executing a message passing +work~\cite{Our_first_paper}, we proposed a method that selects the optimal +frequency scaling factor for a homogeneous cluster executing a message passing iterative synchronous application while giving the best trade-off between the energy consumption and the performance for such applications. In this work we -are interested in heterogeneous grid as described above. Due to the +are interested in heterogeneous clusters as described above. Due to the heterogeneity of the processors, a vector of scaling factors should be selected and it must give the best trade-off between energy consumption and performance. @@ -329,58 +479,49 @@ are not measured using the same metric. To solve this problem, the execution time is normalized by computing the ratio between the new execution time (after scaling down the frequencies of some processors) and the initial one (with maximum frequency for all nodes) as follows: -\begin{equation} +\begin{multline} \label{eq:pnorm} - \Pnorm = \frac{\Tnew}{\Told} -\end{equation} - + \Pnorm = \frac{\Tnew}{\Told}\\ + {} = \frac{ \max_{i=1,2,\dots,N} (\Tcp[i] \cdot S_{i}) +\MinTcm} + {\max_{i=1,2,\dots,N}{(\Tcp[i]+\Tcm[i])}} +\end{multline} -Where $Tnew$ is computed as in (\ref{eq:perf}) and $Told$ is computed as in (\ref{eq:told}) -\begin{equation} - \label{eq:told} - \Told = \mathop{\max_{i=1,2,\dots,N}}_{j=1,2,\dots,M} (\Tcp[ij]+\Tcm[ij]) -\end{equation} In the same way, the energy is normalized by computing the ratio between the consumed energy while scaling down the frequency and the consumed energy with -maximum frequency for all nodes: -\begin{equation} +maximum frequency for all nodes: +\begin{multline} \label{eq:enorm} - \Enorm = \frac{\Ereduced}{\Eoriginal} -\end{equation} - -Where $\Ereduced$ is computed using (\ref{eq:energy}) and $\Eoriginal$ is -computed as in (). - -\textcolor{red}{A reference is missing} -\begin{equation} - \label{eq:eorginal} - \Eoriginal = \sum_{i=1}^{N} \sum_{j=1}^{M} ( \Pd[ij] \cdot \Tcp[ij]) + - \mathop{\sum_{i=1}^{N}} \sum_{j=1}^{M} (\Ps[ij] \cdot \Told) -\end{equation} + \Enorm = \frac{\Ereduced}{\Eoriginal} \\ + {} = \frac{ \sum_{i=1}^{N}{(S_i^{-2} \cdot \Pd[i] \cdot \Tcp[i])} + + \sum_{i=1}^{N} {(\Ps[i] \cdot \Tnew)}}{\sum_{i=1}^{N}{( \Pd[i] \cdot \Tcp[i])} + + \sum_{i=1}^{N} {(\Ps[i] \cdot \Told)}} +\end{multline} +Where $\Ereduced$ and $\Eoriginal$ are computed using (\ref{eq:energy}) and +$\Tnew$ and $\Told$ are computed as in (\ref{eq:pnorm}). While the main goal is to optimize the energy and execution time at the same -time, the normalized energy and execution time curves do not evolve (increase/decrease) in the same way. -According to the equations~(\ref{eq:pnorm}) and (\ref{eq:enorm}), the +time, the normalized energy and execution time curves do not evolve (increase/decrease) in the same way. According to the equations~(\ref{eq:pnorm}) and (\ref{eq:enorm}), the vector of frequency scaling factors $S_1,S_2,\dots,S_N$ reduce both the energy and the execution time simultaneously. But the main objective is to produce maximum energy reduction with minimum execution time reduction. This problem can be solved by making the optimization process for energy and -execution time follow the same evolution according to the vector of scaling factors -$(S_{11}, S_{12},\dots, S_{NM})$. Therefore, the equation of the +execution time follow the same evolution according to the vector of scaling factors. Therefore, the equation of the normalized execution time is inverted which gives the normalized performance equation, as follows: -\begin{equation} +\begin{multline} \label{eq:pnorm_inv} - \Pnorm = \frac{\Told}{\Tnew} -\end{equation} + \Pnorm = \frac{\Told}{\Tnew}\\ + = \frac{\max_{i=1,2,\dots,N}{(\Tcp[i]+\Tcm[i])}} + { \max_{i=1,2,\dots,N} (\Tcp[i] \cdot S_{i}) + \MinTcm} +\end{multline} \begin{figure}[!t] \centering - \subfloat[Homogeneous cluster]{% + \subfloat[Homogeneous platform]{% \includegraphics[width=.33\textwidth]{fig/homo}\label{fig:r1}}% - \subfloat[Heterogeneous grid]{% + \subfloat[Heterogeneous platform]{% \includegraphics[width=.33\textwidth]{fig/heter}\label{fig:r2}} \label{fig:rel} \caption{The energy and performance relation} @@ -395,19 +536,19 @@ Figure~\ref{fig:r2}. Then the objective function has the following form: \begin{equation} \label{eq:max} \MaxDist = -\mathop{ \mathop{\max_{i=1,\dots N}}_{j=1,\dots,M}}_{k=1,\dots,F} - (\overbrace{\Pnorm(S_{ijk})}^{\text{Maximize}} - - \overbrace{\Enorm(S_{ijk})}^{\text{Minimize}} ) + \mathop{\max_{i=1,\dots F}}_{j=1,\dots,N} + (\overbrace{\Pnorm(S_{ij})}^{\text{Maximize}} - + \overbrace{\Enorm(S_{ij})}^{\text{Minimize}} ) \end{equation} -where $N$ is the number of clusters, $M$ is the number of nodes in each cluster and -$F$ is the number of available frequencies for each node. Then, the optimal set -of scaling factors that satisfies (\ref{eq:max}) can be selected. -The objective function can work with any energy model or any power -values for each node (static and dynamic powers). However, the most important -energy reduction gain can be achieved when the energy curve has a convex form as shown +where $N$ is the number of nodes and $F$ is the number of available frequencies +for each node. Then, the optimal set of scaling factors that satisfies +(\ref{eq:max}) can be selected. The objective function can work with any energy +model or any power values for each node (static and dynamic powers). However, +the most important energy reduction gain can be achieved when the energy curve +has a convex form as shown in~\cite{Zhuo_Energy.efficient.Dynamic.Task.Scheduling,Rauber_Analytical.Modeling.for.Energy,Hao_Learning.based.DVFS}. -\section{The scaling factors selection algorithm for grids } +\section{The scaling factors selection algorithm for heterogeneous platforms } \label{sec.optim} \begin{algorithm} @@ -415,44 +556,44 @@ in~\cite{Zhuo_Energy.efficient.Dynamic.Task.Scheduling,Rauber_Analytical.Modelin % \footnotesize \Require ~ \begin{description} - \item [{$N$}] number of clusters in the grid. - \item [{$M$}] number of nodes in each cluster. - \item[{$\Tcp[ij]$}] array of all computation times for all nodes during one iteration and with the highest frequency. - \item[{$\Tcm[ij]$}] array of all communication times for all nodes during one iteration and with the highest frequency. - \item[{$\Fmax[ij]$}] array of the maximum frequencies for all nodes. - \item[{$\Pd[ij]$}] array of the dynamic powers for all nodes. - \item[{$\Ps[ij]$}] array of the static powers for all nodes. - \item[{$\Fdiff[ij]$}] array of the differences between two successive frequencies for all nodes. + \item[{$\Tcp[i]$}] array of all computation times for all nodes during one iteration and with highest frequency. + \item[{$\Tcm[i]$}] array of all communication times for all nodes during one iteration and with highest frequency. + \item[{$\Fmax[i]$}] array of the maximum frequencies for all nodes. + \item[{$\Pd[i]$}] array of the dynamic powers for all nodes. + \item[{$\Ps[i]$}] array of the static powers for all nodes. + \item[{$\Fdiff[i]$}] array of the differences between two successive frequencies for all nodes. \end{description} - \Ensure $\Sopt[11],\Sopt[12] \dots, \Sopt[NM_i]$, a vector of scaling factors that gives the optimal tradeoff between energy consumption and execution time + \Ensure $\Sopt[1],\Sopt[2] \dots, \Sopt[N]$ is a vector of optimal scaling factors - \State $\Scp[ij] \gets \frac{\max_{i=1,2,\dots,N}(\max_{j=1,2,\dots,M_i}(\Tcp[ij]))}{\Tcp[ij]} $ - \State $F_{ij} \gets \frac{\Fmax[ij]}{\Scp[i]},~{i=1,2,\cdots,N},~{j=1,2,\dots,M_i}.$ - \State Round the computed initial frequencies $F_i$ to the closest available frequency for each node. + \State $\Scp[i] \gets \frac{\max_{i=1,2,\dots,N}(\Tcp[i])}{\Tcp[i]} $ + \State $F_{i} \gets \frac{\Fmax[i]}{\Scp[i]},~{i=1,2,\cdots,N}$ + \State Round the computed initial frequencies $F_i$ to the closest one available in each node. \If{(not the first frequency)} - \State $F_{ij} \gets F_{ij}+\Fdiff[ij],~i=1,\dots,N,~{j=1,\dots,M_i}.$ + \State $F_i \gets F_i+\Fdiff[i],~i=1,\dots,N.$ \EndIf - \State $\Told \gets $ computed as in equations (\ref{eq:told}). - \State $\Eoriginal \gets $ computed as in equations (\ref{eq:eorginal}) . - \State $\Sopt[ij] \gets 1,~i=1,\dots,N,~{j=1,\dots,M_i}. $ + \State $\Told \gets \max_{i=1,\dots,N} (\Tcp[i]+\Tcm[i])$ + % \State $\Eoriginal \gets \sum_{i=1}^{N}{( \Pd[i] \cdot \Tcp[i])} +\sum_{i=1}^{N} {(\Ps[i] \cdot \Told)}$ + \State $\Eoriginal \gets \sum_{i=1}^{N}{( \Pd[i] \cdot \Tcp[i] + \Ps[i] \cdot \Told)}$ + \State $\Sopt[i] \gets 1,~i=1,\dots,N. $ \State $\Dist \gets 0 $ - \While {(all nodes have not reached their minimum \newline\hspace*{2.5em} frequency \textbf{or} $\Pnorm - \Enorm < 0 $)} + \While {(all nodes not reach their minimum frequency)} \If{(not the last freq. \textbf{and} not the slowest node)} - \State $F_{ij} \gets F_{ij} - \Fdiff[ij],~{i=1,\dots,N},~{j=1,\dots,M_i}$. - \State $S_{ij} \gets \frac{\Fmax[ij]}{F_{ij}},~{i=1,\dots,N},~{j=1,\dots,M_i}.$ + \State $F_i \gets F_i - \Fdiff[i],~i=1,\dots,N.$ + \State $S_i \gets \frac{\Fmax[i]}{F_i},~i=1,\dots,N.$ \EndIf - \State $\Tnew \gets $ computed as in equations (\ref{eq:perf}). - \State $\Ereduced \gets $ computed as in equations (\ref{eq:energy}). + \State $\Tnew \gets \max_{i=1,\dots,N} (\Tcp[i] \cdot S_{i}) + \MinTcm $ +% \State $\Ereduced \gets \sum_{i=1}^{N}{(S_i^{-2} \cdot \Pd[i] \cdot \Tcp[i])} + \sum_{i=1}^{N} {(\Ps[i] \cdot \rlap{\Tnew)}} $ + \State $\Ereduced \gets \sum_{i=1}^{N}{(S_i^{-2} \cdot \Pd[i] \cdot \Tcp[i] + \Ps[i] \cdot \rlap{\Tnew)}} $ \State $\Pnorm \gets \frac{\Told}{\Tnew}$ \State $\Enorm\gets \frac{\Ereduced}{\Eoriginal}$ \If{$(\Pnorm - \Enorm > \Dist)$} - \State $\Sopt[ij] \gets S_{ij},~i=1,\dots,N,~j=1,\dots,M_i. $ + \State $\Sopt[i] \gets S_{i},~i=1,\dots,N. $ \State $\Dist \gets \Pnorm - \Enorm$ \EndIf \EndWhile - \State Return $\Sopt[11],\Sopt[12],\dots,\Sopt[NM_i]$ + \State Return $\Sopt[1],\Sopt[2],\dots,\Sopt[N]$ \end{algorithmic} - \caption{Scaling factors selection algorithm} + \caption{frequency scaling factors selection algorithm} \label{HSA} \end{algorithm} @@ -478,12 +619,10 @@ in~\cite{Zhuo_Energy.efficient.Dynamic.Task.Scheduling,Rauber_Analytical.Modelin \subsection{The algorithm details} -\textcolor{red}{Delete the subsection if there's only one.} - -In this section, the scaling factors selection algorithm for grids, algorithm~\ref{HSA}, is presented. It selects the vector of the frequency -scaling factors that gives the best trade-off between minimizing the +In this section, Algorithm~\ref{HSA} is presented. It selects the frequency +scaling factors vector that gives the best trade-off between minimizing the energy consumption and maximizing the performance of a message passing -synchronous iterative application executed on a grid. It works +synchronous iterative application executed on a heterogeneous platform. It works online during the execution time of the iterative message passing program. It uses information gathered during the first iteration such as the computation time and the communication time in one iteration for each node. The algorithm is @@ -496,18 +635,18 @@ scaling algorithm is called in the iterative MPI program. \begin{figure}[!t] \centering - \includegraphics[scale=0.45]{fig/init_freq} + \includegraphics[scale=0.5]{fig/start_freq} \caption{Selecting the initial frequencies} \label{fig:st_freq} \end{figure} -Nodes from distinct clusters in a grid have different computing powers, thus +The nodes in a heterogeneous platform have different computing powers, thus while executing message passing iterative synchronous applications, fast nodes have to wait for the slower ones to finish their computations before being able to synchronously communicate with them as in Figure~\ref{fig:heter}. These periods are called idle or slack times. The algorithm takes into account this -problem and tries to reduce these slack times when selecting the vector of the frequency -scaling factors. At first, it selects initial frequency scaling factors +problem and tries to reduce these slack times when selecting the frequency +scaling factors vector. At first, it selects initial frequency scaling factors that increase the execution times of fast nodes and minimize the differences between the computation times of fast and slow nodes. The value of the initial frequency scaling factor for each node is inversely proportional to its @@ -516,7 +655,7 @@ frequency scaling factors are computed as a ratio between the computation time of the slowest node and the computation time of the node $i$ as follows: \begin{equation} \label{eq:Scp} - \Scp[ij] = \frac{ \mathop{\max_{i=1,\dots N}}_{j=1,\dots,M}(\Tcp[ij])} {\Tcp[ij]} + \Scp[i] = \frac{\max_{i=1,2,\dots,N}(\Tcp[i])}{\Tcp[i]} \end{equation} Using the initial frequency scaling factors computed in (\ref{eq:Scp}), the algorithm computes the initial frequencies for all nodes as a ratio between the @@ -524,28 +663,24 @@ maximum frequency of node $i$ and the computation scaling factor $\Scp[i]$ as follows: \begin{equation} \label{eq:Fint} - F_{ij} = \frac{\Fmax[ij]}{\Scp[ij]},~{i=1,2,\dots,N},~{j=1,\dots,M} + F_{i} = \frac{\Fmax[i]}{\Scp[i]},~{i=1,2,\dots,N} \end{equation} If the computed initial frequency for a node is not available in the gears of that node, it is replaced by the nearest available frequency. In -Figure~\ref{fig:st_freq}, the nodes are sorted by their computing powers in +Figure~\ref{fig:st_freq}, the nodes are sorted by their computing power in ascending order and the frequencies of the faster nodes are scaled down according to the computed initial frequency scaling factors. The resulting new frequencies are highlighted in Figure~\ref{fig:st_freq}. This set of frequencies can be considered as a higher bound for the search space of the -optimal vector of frequencies because selecting higher frequencies +optimal vector of frequencies because selecting scaling factors higher than the higher bound will not improve the performance of the application and it will increase its overall energy consumption. Therefore the algorithm that selects the frequency scaling factors starts the search method from these initial frequencies and takes a downward search direction toward lower -frequencies until reaching the nodes' minimum frequencies or lower bounds. A node's frequency is considered its lower bound if the computed distance between the energy and performance at this frequency is less than zero. -A negative distance means that the performance degradation ratio is higher than the energy saving ratio. -In this situation, the algorithm must stop the downward search because it has reached the lower bound and it is useless to test the lower frequencies. Indeed, they will all give worse distances. - -Therefore, the algorithm iterates on all remaining frequencies, from the higher -bound until all nodes reach their minimum frequencies or their lower bounds, to compute the overall -energy consumption and performance and selects the optimal vector of the frequency scaling -factors. At each iteration the algorithm determines the slowest node +frequencies. The algorithm iterates on all remaining frequencies, from the higher +bound until all nodes reach their minimum frequencies, to compute their overall +energy consumption and performance, and select the optimal frequency scaling +factors vector. At each iteration the algorithm determines the slowest node according to the equation (\ref{eq:perf}) and keeps its frequency unchanged, while it lowers the frequency of all other nodes by one gear. The new overall energy consumption and execution time are computed according to the new scaling @@ -553,147 +688,541 @@ factors. The optimal set of frequency scaling factors is the set that gives the highest distance according to the objective function (\ref{eq:max}). Figures~\ref{fig:r1} and \ref{fig:r2} illustrate the normalized performance and -consumed energy for an application running on a homogeneous cluster and a - grid platform respectively while increasing the scaling factors. It can -be noticed that in a homogeneous cluster the search for the optimal scaling +consumed energy for an application running on a homogeneous platform and a +heterogeneous platform respectively while increasing the scaling factors. It can +be noticed that in a homogeneous platform the search for the optimal scaling factor should start from the maximum frequency because the performance and the consumed energy decrease from the beginning of the plot. On the other hand, in -the grid platform the performance is maintained at the beginning of the +the heterogeneous platform the performance is maintained at the beginning of the plot even if the frequencies of the faster nodes decrease until the computing power of scaled down nodes are lower than the slowest node. In other words, until they reach the higher bound. It can also be noticed that the higher the difference between the faster nodes and the slower nodes is, the bigger the -maximum distance between the energy curve and the performance curve is, which results in bigger energy savings. +maximum distance between the energy curve and the performance curve is while the +scaling factors are varying which results in bigger energy savings. +Finally, in a homogeneous platform the energy consumption is increased when the scaling factor is very high. +Indeed, the dynamic energy saved by reducing the frequency of the processor is compensated by the significant increase of the execution time and thus the increased of the static energy. On the other hand, in a heterogeneous platform this is not the case. + +\subsection{The evaluation of the proposed algorithm} +\label{sec.verif.algo} + +The precision of the proposed algorithm mainly depends on the execution time +prediction model defined in (\ref{eq:perf}) and the energy model computed by +(\ref{eq:energy}). The energy model is also significantly dependent on the +execution time model because the static energy is linearly related to the +execution time and the dynamic energy is related to the computation time. So, +all the works presented in this paper are based on the execution time model. To +verify this model, the predicted execution time was compared to the real +execution time over SimGrid/SMPI simulator, +v3.10~\cite{casanova+giersch+legrand+al.2014.versatile}, for all the NAS +parallel benchmarks NPB v3.3 \cite{NAS.Parallel.Benchmarks}, running class B on +8 or 9 nodes. The comparison showed that the proposed execution time model is +very precise, the maximum normalized difference between the predicted execution +time and the real execution time is equal to 0.03 for all the NAS benchmarks. + +Since the proposed algorithm is not an exact method, it does not test all the +possible solutions (vectors of scaling factors) in the search space. To prove +its efficiency, it was compared on small instances to a brute force search +algorithm that tests all the possible solutions. The brute force algorithm was +applied to different NAS benchmarks classes with different number of nodes. The +solutions returned by the brute force algorithm and the proposed algorithm were +identical and the proposed algorithm was on average 10 times faster than the +brute force algorithm. It has a small execution time: for a heterogeneous +cluster composed of four different types of nodes having the characteristics +presented in Table~\ref{table:platform}, it takes on average \np[ms]{0.04} for 4 +nodes and \np[ms]{0.15} on average for 144 nodes to compute the best scaling +factors vector. The algorithm complexity is $O(F\cdot N)$, where $F$ is the +maximum number of available frequencies, and $N$ is the number of computing +nodes. The algorithm needs from 12 to 20 iterations to select the best vector of +frequency scaling factors that gives the results of the next sections. +\begin{table}[!t] + \caption{Heterogeneous nodes characteristics} + % title of Table + \centering + \begin{tabular}{|*{7}{r|}} + \hline + Node & Simulated & Max & Min & Diff. & Dynamic & Static \\ + type & GFLOPS & Freq. & Freq. & Freq. & power & power \\ + & & GHz & GHz & GHz & & \\ + \hline + 1 & 40 & 2.50 & 1.20 & 0.100 & \np[W]{20} & \np[W]{4} \\ + \hline + 2 & 50 & 2.66 & 1.60 & 0.133 & \np[W]{25} & \np[W]{5} \\ + \hline + 3 & 60 & 2.90 & 1.20 & 0.100 & \np[W]{30} & \np[W]{6} \\ + \hline + 4 & 70 & 3.40 & 1.60 & 0.133 & \np[W]{35} & \np[W]{7} \\ + \hline + \end{tabular} + \label{table:platform} +\end{table} \section{Experimental results} \label{sec.expe} -While in~\cite{mpi-energy2} the energy model and the scaling factors selection algorithm were applied to a heterogeneous cluster and evaluated over the SimGrid simulator~\cite{SimGrid.org}, -in this paper real experiments were conducted over the grid'5000 platform. - -\subsection{Grid'5000 architature and power consumption} -\label{sec.grid5000} -Grid'5000~\cite{grid5000} is a large-scale testbed that consists of ten sites distributed over all metropolitan France and Luxembourg. All the sites are connected together via a special long distance network called RENATER, -which is the French National Telecommunication Network for Technology. -Each site of the grid is composed of few heterogeneous -computing clusters and each cluster contains many homogeneous nodes. In total, - grid'5000 has about one thousand heterogeneous nodes and eight thousand cores. In each site, -the clusters and their nodes are connected via high speed local area networks. -Two types of local networks are used, Ethernet or Infiniband networks which have different characteristics in terms of bandwidth and latency. - -Since grid'5000 is dedicated for testing, contrary to production grids it allows a user to deploy its own customized operating system on all the booked nodes. The user could have root rights and thus apply DVFS operations while executing a distributed application. Moreover, the grid'5000 testbed provides at some sites a power measurement tool to capture -the power consumption for each node in those sites. The measured power is the overall consumed power by by all the components of a node at a given instant, such as CPU, hard drive, main-board, memory, ... For more details refer to -\cite{Energy_measurement}. To just measure the CPU power of one core in a node $j$, - firstly, the power consumed by the node while being idle at instant $y$, noted as $\Pidle[jy]$, was measured. Then, the power was measured while running a single thread benchmark with no communication (no idle time) over the same node with its CPU scaled to the maximum available frequency. The latter power measured at time $x$ with maximum frequency for one core of node $j$ is noted $P\max[jx]$. The difference between the two measured power consumption represents the -dynamic power consumption of that core with the maximum frequency, see figure(\ref{fig:power_cons}). - -\textcolor{red}{why maximum and minimum, change peak in the equation and the figure} - -The dynamic power $\Pd[j]$ is computed as in equation (\ref{eq:pdyn}) -\begin{equation} - \label{eq:pdyn} - \Pd[j] = \max_{x=\beta_1,\dots \beta_2} (P\max[jx]) - \min_{y=\Theta_1,\dots \Theta_2} (\Pidle[jy]) -\end{equation} - -where $\Pd[j]$ is the dynamic power consumption for one core of node $j$, -$\lbrace \beta_1,\beta_2 \rbrace$ is the time interval for the measured peak power values, -$\lbrace\Theta_1,\Theta_2\rbrace$ is the time interval for the measured idle power values. -Therefore, the dynamic power of one core is computed as the difference between the maximum -measured value in peak powers vector and the minimum measured value in the idle powers vector. -On the other hand, the static power consumption by one core is a part of the measured idle power consumption of the node. Since in grid'5000 there is no way to measure precisely the consumed static power and in~\cite{Our_first_paper,pdsec2015,Rauber_Analytical.Modeling.for.Energy} it was assumed that the static power represents a ratio of the dynamic power, the value of the static power is assumed as np[\%]{20} of dynamic power consumption of the core. +To evaluate the efficiency and the overall energy consumption reduction of +Algorithm~\ref{HSA}, it was applied to the NAS parallel benchmarks NPB v3.3 which +is composed of synchronous message passing applications. The +experiments were executed on the simulator SimGrid/SMPI which offers easy tools +to create a heterogeneous platform and run message passing applications over it. +The heterogeneous platform that was used in the experiments, had one core per +node because just one process was executed per node. The heterogeneous platform +was composed of four types of nodes. Each type of nodes had different +characteristics such as the maximum CPU frequency, the number of available +frequencies and the computational power, see Table~\ref{table:platform}. The +characteristics of these different types of nodes are inspired from the +specifications of real Intel processors. The heterogeneous platform had up to +144 nodes and had nodes from the four types in equal proportions, for example if +a benchmark was executed on 8 nodes, 2 nodes from each type were used. Since the +constructors of CPUs do not specify the dynamic and the static power of their +CPUs, for each type of node they were chosen proportionally to its computing +power (FLOPS). In the initial heterogeneous platform, while computing with +highest frequency, each node consumed an amount of power proportional to its +computing power (which corresponds to \np[\%]{80} of its dynamic power and the +remaining \np[\%]{20} to the static power), the same assumption was made in +\cite{Our_first_paper,Rauber_Analytical.Modeling.for.Energy}. Finally, These +nodes were connected via an Ethernet network with \np[Gbit/s]{1} bandwidth. -In the experiments presented in the following sections, two sites of grid'5000 were used, Lyon and Nancy sites. These two sites have in total seven different clusters as in figure (\ref{fig:grid5000}). - -Four clusters from the two sites were selected in the experiments: one cluster from -Lyon's site, Taurus cluster, and three clusters from Nancy's site, Graphene, -Griffon and Graphite. Each one of these clusters has homogeneous nodes inside, while nodes from different clusters are heterogeneous in many aspects such as: computing power, power consumption, available -frequency ranges and local network features: the bandwidth and the latency. Table \ref{table:grid5000} shows -the details characteristics of these four clusters. Moreover, the dynamic powers were computed using the equation (\ref{eq:pdyn}) for all the nodes in the -selected clusters and are presented in table \ref{table:grid5000}. +\subsection{The experimental results of the scaling algorithm} +\label{sec.res} +The proposed algorithm was applied to the seven parallel NAS benchmarks (EP, CG, +MG, FT, BT, LU and SP) and the benchmarks were executed with the three classes: +A, B and C. However, due to the lack of space in this paper, only the results of +the biggest class, C, are presented while being run on different number of +nodes, ranging from 8 to 128 or 144 nodes depending on the benchmark being +executed. Indeed, the benchmarks CG, MG, LU, EP and FT had to be executed on 1, +2, 4, 8, 16, 32, 64, or 128 nodes. The other benchmarks such as BT and SP had +to be executed on 1, 4, 9, 16, 36, 64, or 144 nodes. +\begin{table}[!t] + +% \end{table} -\begin{figure}[!t] + +% \begin{table}[!t] + \caption{Running NAS benchmarks on 8 and 9 nodes } + % title of Table \centering - \includegraphics[scale=1]{fig/grid5000} - \caption{The selected two sites of grid'5000} - \label{fig:grid5000} -\end{figure} - + \begin{tabular}{|*{7}{r|}} + \hline + \hspace{-2.2084pt}% + Program & Execution & Energy & Energy & Performance & Distance \\ + name & time/s & consumption/J & saving\% & degradation\% & \\ + \hline + CG & 36.11 & 3263.49 & 31.25 & 7.12 & 24.13 \\ + \hline + MG & 8.99 & 953.39 & 33.78 & 6.41 & 27.37 \\ + \hline + EP & 40.39 & 5652.81 & 27.04 & 0.49 & 26.55 \\ + \hline + LU & 218.79 & 36149.77 & 28.23 & 0.01 & 28.22 \\ + \hline + BT & 166.89 & 23207.42 & 32.32 & 7.89 & 24.43 \\ + \hline + SP & 104.73 & 18414.62 & 24.73 & 2.78 & 21.95 \\ + \hline + FT & 51.10 & 4913.26 & 31.02 & 2.54 & 28.48 \\ + \hline + \end{tabular} + \label{table:res_8n} +% \end{table} -The energy model and the scaling factors selection algorithm were applied to the NAS parallel benchmarks v3.3 \cite{NAS.Parallel.Benchmarks} and evaluated over grid'5000. -The benchmark suite contains seven applications: CG, MG, EP, LU, BT, SP and FT. These applications have different computations and communications ratios and strategies which make them good testbed applications to evaluate the proposed algorithm and energy model. -The benchmarks have seven different classes, S, W, A, B, C, D and E, that represent the size of the problem that the method solves. In this work, the class D was used for all benchmarks in all the experiments presented in the next sections. + \medskip +% \begin{table}[!t] + \caption{Running NAS benchmarks on 16 nodes } + % title of Table + \centering + \begin{tabular}{|*{7}{r|}} + \hline + \hspace{-2.2084pt}% + Program & Execution & Energy & Energy & Performance & Distance \\ + name & time/s & consumption/J & saving\% & degradation\% & \\ + \hline + CG & 31.74 & 4373.90 & 26.29 & 9.57 & 16.72 \\ + \hline + MG & 5.71 & 1076.19 & 32.49 & 6.05 & 26.44 \\ + \hline + EP & 20.11 & 5638.49 & 26.85 & 0.56 & 26.29 \\ + \hline + LU & 144.13 & 42529.06 & 28.80 & 6.56 & 22.24 \\ + \hline + BT & 97.29 & 22813.86 & 34.95 & 5.80 & 29.15 \\ + \hline + SP & 66.49 & 20821.67 & 22.49 & 3.82 & 18.67 \\ + \hline + FT & 37.01 & 5505.60 & 31.59 & 6.48 & 25.11 \\ + \hline + \end{tabular} + \label{table:res_16n} +% \end{table} + \medskip +% \begin{table}[!t] + \caption{Running NAS benchmarks on 32 and 36 nodes } + % title of Table + \centering + \begin{tabular}{|*{7}{r|}} + \hline + \hspace{-2.2084pt}% + Program & Execution & Energy & Energy & Performance & Distance \\ + name & time/s & consumption/J & saving\% & degradation\% & \\ + \hline + CG & 32.35 & 6704.21 & 16.15 & 5.30 & 10.85 \\ + \hline + MG & 4.30 & 1355.58 & 28.93 & 8.85 & 20.08 \\ + \hline + EP & 9.96 & 5519.68 & 26.98 & 0.02 & 26.96 \\ + \hline + LU & 99.93 & 67463.43 & 23.60 & 2.45 & 21.15 \\ + \hline + BT & 48.61 & 23796.97 & 34.62 & 5.83 & 28.79 \\ + \hline + SP & 46.01 & 27007.43 & 22.72 & 3.45 & 19.27 \\ + \hline + FT & 28.06 & 7142.69 & 23.09 & 2.90 & 20.19 \\ + \hline + \end{tabular} + \label{table:res_32n} +% \end{table} + \medskip +% \begin{table}[!t] + \caption{Running NAS benchmarks on 64 nodes } + % title of Table + \centering + \begin{tabular}{|*{7}{r|}} + \hline + \hspace{-2.2084pt}% + Program & Execution & Energy & Energy & Performance & Distance \\ + name & time/s & consumption/J & saving\% & degradation\% & \\ + \hline + CG & 46.65 & 17521.83 & 8.13 & 1.68 & 6.45 \\ + \hline + MG & 3.27 & 1534.70 & 29.27 & 14.35 & 14.92 \\ + \hline + EP & 5.05 & 5471.11 & 27.12 & 3.11 & 24.01 \\ + \hline + LU & 73.92 & 101339.16 & 21.96 & 3.67 & 18.29 \\ + \hline + BT & 39.99 & 27166.71 & 32.02 & 12.28 & 19.74 \\ + \hline + SP & 52.00 & 49099.28 & 24.84 & 0.03 & 24.81 \\ + \hline + FT & 25.97 & 10416.82 & 20.15 & 4.87 & 15.28 \\ + \hline + \end{tabular} + \label{table:res_64n} +% \end{table} + \medskip +% \begin{table}[!t] + \caption{Running NAS benchmarks on 128 and 144 nodes } + % title of Table + \centering + \begin{tabular}{|*{7}{r|}} + \hline + \hspace{-2.2084pt}% + Program & Execution & Energy & Energy & Performance & Distance \\ + name & time/s & consumption/J & saving\% & degradation\% & \\ + \hline + CG & 56.92 & 41163.36 & 4.00 & 1.10 & 2.90 \\ + \hline + MG & 3.55 & 2843.33 & 18.77 & 10.38 & 8.39 \\ + \hline + EP & 2.67 & 5669.66 & 27.09 & 0.03 & 27.06 \\ + \hline + LU & 51.23 & 144471.90 & 16.67 & 2.36 & 14.31 \\ + \hline + BT & 37.96 & 44243.82 & 23.18 & 1.28 & 21.90 \\ + \hline + SP & 64.53 & 115409.71 & 26.72 & 0.05 & 26.67 \\ + \hline + FT & 25.51 & 18808.72 & 12.85 & 2.84 & 10.01 \\ + \hline + \end{tabular} + \label{table:res_128n} +\end{table} \begin{figure}[!t] \centering - \includegraphics[scale=0.6]{fig/power_consumption.pdf} - \caption{The power consumption by one core from Taurus cluster} - \label{fig:power_cons} + \subfloat[Energy saving (\%)]{% + \includegraphics[width=.33\textwidth]{fig/energy}\label{fig:energy}}% + + \subfloat[Performance degradation (\%)]{% + \includegraphics[width=.33\textwidth]{fig/per_deg}\label{fig:per_deg}} + \label{fig:avg} + \caption{The energy and performance for all NAS benchmarks running with a different number of nodes} \end{figure} +The overall energy consumption was computed for each instance according to the +energy consumption model (\ref{eq:energy}), with and without applying the +algorithm. The execution time was also measured for all these experiments. Then, +the energy saving and performance degradation percentages were computed for each +instance. The results are presented in Tables +\ref{table:res_8n}, \ref{table:res_16n}, \ref{table:res_32n}, +\ref{table:res_64n} and \ref{table:res_128n}. All these results are the average +values from many experiments for energy savings and performance degradation. +The tables show the experimental results for running the NAS parallel benchmarks +on different numbers of nodes. The experiments show that the algorithm +significantly reduces the energy consumption (up to \np[\%]{34}) and tries to +limit the performance degradation. They also show that the energy saving +percentage decreases when the number of computing nodes increases. This +reduction is due to the increase of the communication times compared to the +execution times when the benchmarks are run over a higher number of nodes. +Indeed, the benchmarks with the same class, C, are executed on different numbers +of nodes, so the computation required for each iteration is divided by the +number of computing nodes. On the other hand, more communications are required +when increasing the number of nodes so the static energy increases linearly +according to the communication time and the dynamic power is less relevant in +the overall energy consumption. Therefore, reducing the frequency with +Algorithm~\ref{HSA} is less effective in reducing the overall energy savings. It +can also be noticed that for the benchmarks EP and SP that contain little or no +communications, the energy savings are not significantly affected by the high +number of nodes. No experiments were conducted using bigger classes than D, +because they require a lot of memory (more than \np[GB]{64}) when being executed +by the simulator on one machine. The maximum distance between the normalized +energy curve and the normalized performance for each instance is also shown in +the result tables. It decrease in the same way as the energy saving percentage. +The tables also show that the performance degradation percentage is not +significantly increased when the number of computing nodes is increased because +the computation times are small when compared to the communication times. + +Figures~\ref{fig:energy} and \ref{fig:per_deg} present the energy saving and +performance degradation respectively for all the benchmarks according to the +number of used nodes. As shown in the first plot, the energy saving percentages +of the benchmarks MG, LU, BT and FT decrease linearly when the number of nodes +increase. While for the EP and SP benchmarks, the energy saving percentage is +not affected by the increase of the number of computing nodes, because in these +benchmarks there are little or no communications. Finally, the energy saving of +the CG benchmark significantly decreases when the number of nodes increase +because this benchmark has more communications than the others. The second plot +shows that the performance degradation percentages of most of the benchmarks +decrease when they run on a big number of nodes because they spend more time +communicating than computing, thus, scaling down the frequencies of some nodes +has less effect on the performance. +\subsection{The results for different power consumption scenarios} +\label{sec.compare} + +The results of the previous section were obtained while using processors that +consume during computation an overall power which is \np[\%]{80} composed of +dynamic power and of \np[\%]{20} of static power. In this section, these ratios +are changed and two new power scenarios are considered in order to evaluate how +the proposed algorithm adapts itself according to the static and dynamic power +values. The two new power scenarios are the following: + +\begin{itemize} +\item \np[\%]{70} of dynamic power and \np[\%]{30} of static power +\item \np[\%]{90} of dynamic power and \np[\%]{10} of static power +\end{itemize} + +The NAS parallel benchmarks were executed again over processors that follow the +new power scenarios. The class C of each benchmark was run over 8 or 9 nodes +and the results are presented in Tables~\ref{table:res_s1} and +\ref{table:res_s2}. These tables show that the energy saving percentage of the +\np[\%]{70}-\np[\%]{30} scenario is smaller for all benchmarks compared to the +energy saving of the \np[\%]{90}-\np[\%]{10} scenario. Indeed, in the latter +more dynamic power is consumed when nodes are running on their maximum +frequencies, thus, scaling down the frequency of the nodes results in higher +energy savings than in the \np[\%]{70}-\np[\%]{30} scenario. On the other hand, +the performance degradation percentage is smaller in the \np[\%]{70}-\np[\%]{30} +scenario compared to the \np[\%]{90}-\np[\%]{10} scenario. This is due to the +higher static power percentage in the first scenario which makes it more +relevant in the overall consumed energy. Indeed, the static energy is related +to the execution time and if the performance is degraded the amount of consumed +static energy directly increases. Therefore, the proposed algorithm does not +really significantly scale down much the frequencies of the nodes in order to +limit the increase of the execution time and thus limiting the effect of the +consumed static energy. + +Both new power scenarios are compared to the old one in +Figure~\ref{fig:sen_comp}. It shows the average of the performance degradation, +the energy saving and the distances for all NAS benchmarks of class C running on +8 or 9 nodes. The comparison shows that the energy saving ratio is proportional +to the dynamic power ratio: it is increased when applying the +\np[\%]{90}-\np[\%]{10} scenario because at maximum frequency the dynamic energy +is the most relevant in the overall consumed energy and can be reduced by +lowering the frequency of some processors. On the other hand, the energy saving +decreases when the \np[\%]{70}-\np[\%]{30} scenario is used because the dynamic +energy is less relevant in the overall consumed energy and lowering the +frequency does not return big energy savings. Moreover, the average of the +performance degradation is decreased when using a higher ratio for static power +(e.g. \np[\%]{70}-\np[\%]{30} scenario and \np[\%]{80}-\np[\%]{20} +scenario). Since the proposed algorithm optimizes the energy consumption when +using a higher ratio for dynamic power the algorithm selects bigger frequency +scaling factors that result in more energy saving but less performance, for +example see Figure~\ref{fig:scales_comp}. The opposite happens when using a +higher ratio for static power, the algorithm proportionally selects smaller +scaling values which result in less energy saving but also less performance +degradation. - \begin{table}[!t] - \caption{CPUs characteristics of the selected clusters} + \caption{The results of the \np[\%]{70}-\np[\%]{30} power scenario} % title of Table \centering - \begin{tabular}{|*{7}{c|}} + \begin{tabular}{|*{6}{r|}} \hline - Cluster & CPU & Max & Min & Diff. & no. of cores & dynamic power \\ - Name & model & Freq. & Freq. & Freq. & per CPU & of one core \\ - & & GHz & GHz & GHz & & \\ + Program & Energy & Energy & Performance & Distance \\ + name & consumption/J & saving\% & degradation\% & \\ \hline - Taurus & Intel & 2.3 & 1.2 & 0.1 & 6 & \np[W]{35} \\ - & Xeon & & & & & \\ - & E5-2630 & & & & & \\ + CG & 4144.21 & 22.42 & 7.72 & 14.70 \\ \hline - Graphene & Intel & 2.53 & 1.2 & 0.133 & 4 & \np[W]{23} \\ - & Xeon & & & & & \\ - & X3440 & & & & & \\ + MG & 1133.23 & 24.50 & 5.34 & 19.16 \\ + \hline + EP & 6170.30 & 16.19 & 0.02 & 16.17 \\ + \hline + LU & 39477.28 & 20.43 & 0.07 & 20.36 \\ \hline - Griffon & Intel & 2.5 & 2 & 0.5 & 4 & \np[W]{46} \\ - & Xeon & & & & & \\ - & L5420 & & & & & \\ - \hline - Graphite & Intel & 2 & 1.2 & 0.1 & 8 & \np[W]{35} \\ - & Xeon & & & & & \\ - & E5-2650 & & & & & \\ + BT & 26169.55 & 25.34 & 6.62 & 18.71 \\ + \hline + SP & 19620.09 & 19.32 & 3.66 & 15.66 \\ + \hline + FT & 6094.07 & 23.17 & 0.36 & 22.81 \\ \hline \end{tabular} - \label{table:grid5000} -\end{table} - - - + \label{table:res_s1} +\end{table} -\subsection{The experimental results of the scaling algorithm} -\label{sec.res} +\begin{table}[!t] + \caption{The results of the \np[\%]{90}-\np[\%]{10} power scenario} + % title of Table + \centering + \begin{tabular}{|*{6}{r|}} + \hline + Program & Energy & Energy & Performance & Distance \\ + name & consumption/J & saving\% & degradation\% & \\ + \hline + CG & 2812.38 & 36.36 & 6.80 & 29.56 \\ + \hline + MG & 825.43 & 38.35 & 6.41 & 31.94 \\ + \hline + EP & 5281.62 & 35.02 & 2.68 & 32.34 \\ + \hline + LU & 31611.28 & 39.15 & 3.51 & 35.64 \\ + \hline + BT & 21296.46 & 36.70 & 6.60 & 30.10 \\ + \hline + SP & 15183.42 & 35.19 & 11.76 & 23.43 \\ + \hline + FT & 3856.54 & 40.80 & 5.67 & 35.13 \\ + \hline + \end{tabular} + \label{table:res_s2} +\end{table} -\subsection{The experimental results of multi-cores clusters} -\label{sec.res} +\begin{table}[!t] + \caption{Comparing the proposed algorithm} + \centering + \begin{tabular}{|*{7}{r|}} + \hline + Program & \multicolumn{2}{c|}{Energy saving \%} + & \multicolumn{2}{c|}{Perf. degradation \%} + & \multicolumn{2}{c|}{Distance} \\ + \cline{2-7} + name & EDP & MaxDist & EDP & MaxDist & EDP & MaxDist \\ + \hline + CG & 27.58 & 31.25 & 5.82 & 7.12 & 21.76 & 24.13 \\ + \hline + MG & 29.49 & 33.78 & 3.74 & 6.41 & 25.75 & 27.37 \\ + \hline + LU & 19.55 & 28.33 & 0.00 & 0.01 & 19.55 & 28.22 \\ + \hline + EP & 28.40 & 27.04 & 4.29 & 0.49 & 24.11 & 26.55 \\ + \hline + BT & 27.68 & 32.32 & 6.45 & 7.87 & 21.23 & 24.43 \\ + \hline + SP & 20.52 & 24.73 & 5.21 & 2.78 & 15.31 & 21.95 \\ + \hline + FT & 27.03 & 31.02 & 2.75 & 2.54 & 24.28 & 28.48 \\ + \hline + \end{tabular} + \label{table:compare_EDP} +\end{table} -\subsection{The results for different power consumption scenarios} -\label{sec.compare} +\begin{figure}[!t] + \centering + \subfloat[Comparison between the results on 8 nodes]{% + \includegraphics[width=.33\textwidth]{fig/sen_comp}\label{fig:sen_comp}}% + \subfloat[Comparison the selected frequency scaling factors of MG benchmark class C running on 8 nodes]{% + \includegraphics[width=.33\textwidth]{fig/three_scenarios}\label{fig:scales_comp}} + \label{fig:comp} + \caption{The comparison of the three power scenarios} +\end{figure} +\begin{figure}[!t] + \centering + \includegraphics[scale=0.5]{fig/compare_EDP.pdf} + \caption{Trade-off comparison for NAS benchmarks class C} + \label{fig:compare_EDP} +\end{figure} \subsection{The comparison of the proposed scaling algorithm } \label{sec.compare_EDP} - +In this section, the scaling factors selection algorithm, called MaxDist, is +compared to Spiliopoulos et al. algorithm +\cite{Spiliopoulos_Green.governors.Adaptive.DVFS}, called EDP. They developed a +green governor that regularly applies an online frequency selecting algorithm to +reduce the energy consumed by a multicore architecture without degrading much +its performance. The algorithm selects the frequencies that minimize the energy +and delay products, $\mathit{EDP}=\mathit{energy}\times \mathit{delay}$ using +the predicted overall energy consumption and execution time delay for each +frequency. To fairly compare both algorithms, the same energy and execution +time models, equations (\ref{eq:energy}) and (\ref{eq:fnew}), were used for both +algorithms to predict the energy consumption and the execution times. Also +Spiliopoulos et al. algorithm was adapted to start the search from the initial +frequencies computed using the equation (\ref{eq:Fint}). The resulting algorithm +is an exhaustive search algorithm that minimizes the EDP and has the initial +frequencies values as an upper bound. + +Both algorithms were applied to the parallel NAS benchmarks to compare their +efficiency. Table~\ref{table:compare_EDP} presents the results of comparing the +execution times and the energy consumption for both versions of the NAS +benchmarks while running the class C of each benchmark over 8 or 9 heterogeneous +nodes. The results show that our algorithm provides better energy savings than +Spiliopoulos et al. algorithm, on average it results in \np[\%]{29.76} energy +saving while their algorithm returns just \np[\%]{25.75}. The average of +performance degradation percentage is approximately the same for both +algorithms, about \np[\%]{4}. + +For all benchmarks, our algorithm outperforms Spiliopoulos et al. algorithm in +terms of energy and performance trade-off, see Figure~\ref{fig:compare_EDP}, +because it maximizes the distance between the energy saving and the performance +degradation values while giving the same weight for both metrics. \section{Conclusion} \label{sec.concl} - +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 trade-off) 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 platforms. 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 \np[\%]{34} 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. Finally, the algorithm was +compared to Spiliopoulos et al. algorithm and the results showed that it +outperforms their algorithm in terms of energy-time trade-off. + +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 platforms 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 other tasks to finish their works. The +development of such a method might require a new energy model because the number +of iterations is not known in advance and depends on the global convergence of +the iterative system. \section*{Acknowledgment} diff --git a/fig/commtasks.pdf b/fig/commtasks.pdf index 1b1498c..0e39a6d 100644 Binary files a/fig/commtasks.pdf and b/fig/commtasks.pdf differ diff --git a/fig/compare_EDP.pdf b/fig/compare_EDP.pdf new file mode 100644 index 0000000..62b710d Binary files /dev/null and b/fig/compare_EDP.pdf differ diff --git a/fig/energy.pdf b/fig/energy-eps-converted-to.pdf similarity index 95% rename from fig/energy.pdf rename to fig/energy-eps-converted-to.pdf index f16f58a..11bffb7 100644 Binary files a/fig/energy.pdf and b/fig/energy-eps-converted-to.pdf differ diff --git a/fig/energy.eps b/fig/energy.eps new file mode 100644 index 0000000..2a54918 --- /dev/null +++ b/fig/energy.eps @@ -0,0 +1,868 @@ +%!PS-Adobe-2.0 EPSF-2.0 +%%Title: energy.eps +%%Creator: gnuplot 4.6 patchlevel 0 +%%CreationDate: Thu Feb 19 16:44:03 2015 +%%DocumentFonts: (atend) +%%BoundingBox: 50 50 320 239 +%%EndComments +%%BeginProlog +/gnudict 256 dict def +gnudict begin +% +% The following true/false flags may be edited by hand if desired. +% The unit line width and grayscale image gamma correction may also be changed. +% +/Color false def +/Blacktext false def +/Solid false def +/Dashlength 1 def +/Landscape false def +/Level1 false def +/Rounded false def +/ClipToBoundingBox false def +/SuppressPDFMark false def +/TransparentPatterns false def +/gnulinewidth 5.000 def +/userlinewidth gnulinewidth def +/Gamma 1.0 def +/BackgroundColor {-1.000 -1.000 -1.000} def +% +/vshift -46 def +/dl1 { + 10.0 Dashlength mul mul + Rounded { currentlinewidth 0.75 mul sub dup 0 le { pop 0.01 } if } if +} def +/dl2 { + 10.0 Dashlength mul mul + Rounded { currentlinewidth 0.75 mul add } if +} def +/hpt_ 31.5 def +/vpt_ 31.5 def +/hpt hpt_ def +/vpt vpt_ def +/doclip { + ClipToBoundingBox { + newpath 50 50 moveto 320 50 lineto 320 239 lineto 50 239 lineto closepath + clip + } if +} def +% +% Gnuplot Prolog Version 4.4 (August 2010) +% +%/SuppressPDFMark true def +% +/M {moveto} bind def +/L {lineto} bind def +/R {rmoveto} bind def +/V {rlineto} bind def +/N {newpath moveto} bind def +/Z {closepath} bind def +/C {setrgbcolor} bind def +/f {rlineto fill} bind def +/g {setgray} bind def +/Gshow {show} def % May be redefined later in the file to support UTF-8 +/vpt2 vpt 2 mul def +/hpt2 hpt 2 mul def +/Lshow {currentpoint stroke M 0 vshift R + Blacktext {gsave 0 setgray show grestore} {show} ifelse} def +/Rshow {currentpoint stroke M dup stringwidth pop neg vshift R + Blacktext {gsave 0 setgray show grestore} {show} ifelse} def +/Cshow {currentpoint stroke M dup stringwidth pop -2 div vshift R + Blacktext {gsave 0 setgray show grestore} {show} ifelse} def +/UP {dup vpt_ mul /vpt exch def hpt_ mul /hpt exch def + /hpt2 hpt 2 mul def /vpt2 vpt 2 mul def} def +/DL {Color {setrgbcolor Solid {pop []} if 0 setdash} + {pop pop pop 0 setgray Solid {pop []} if 0 setdash} ifelse} def +/BL {stroke userlinewidth 2 mul setlinewidth + Rounded {1 setlinejoin 1 setlinecap} if} def +/AL {stroke userlinewidth 2 div setlinewidth + Rounded {1 setlinejoin 1 setlinecap} if} def +/UL {dup gnulinewidth mul /userlinewidth exch def + dup 1 lt {pop 1} if 10 mul /udl exch def} def +/PL {stroke userlinewidth setlinewidth + Rounded {1 setlinejoin 1 setlinecap} if} def +3.8 setmiterlimit +% Default Line colors +/LCw {1 1 1} def +/LCb {0 0 0} def +/LCa {0 0 0} def +/LC0 {1 0 0} def +/LC1 {0 1 0} def +/LC2 {0 0 1} def +/LC3 {1 0 1} def +/LC4 {0 1 1} def +/LC5 {1 1 0} def +/LC6 {0 0 0} def +/LC7 {1 0.3 0} def +/LC8 {0.5 0.5 0.5} def +% Default Line Types +/LTw {PL [] 1 setgray} def +/LTb {BL [] LCb DL} def +/LTa {AL [1 udl mul 2 udl mul] 0 setdash LCa setrgbcolor} def +/LT0 {PL [] LC0 DL} def +/LT1 {PL [4 dl1 2 dl2] LC1 DL} def +/LT2 {PL [2 dl1 3 dl2] LC2 DL} def +/LT3 {PL [1 dl1 1.5 dl2] LC3 DL} def +/LT4 {PL [6 dl1 2 dl2 1 dl1 2 dl2] LC4 DL} def +/LT5 {PL [3 dl1 3 dl2 1 dl1 3 dl2] LC5 DL} def +/LT6 {PL [2 dl1 2 dl2 2 dl1 6 dl2] LC6 DL} def +/LT7 {PL [1 dl1 2 dl2 6 dl1 2 dl2 1 dl1 2 dl2] LC7 DL} def +/LT8 {PL [2 dl1 2 dl2 2 dl1 2 dl2 2 dl1 2 dl2 2 dl1 4 dl2] LC8 DL} def +/Pnt {stroke [] 0 setdash gsave 1 setlinecap M 0 0 V stroke grestore} def +/Dia {stroke [] 0 setdash 2 copy vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V closepath stroke + Pnt} def +/Pls {stroke [] 0 setdash vpt sub M 0 vpt2 V + currentpoint stroke M + hpt neg vpt neg R hpt2 0 V stroke + } def +/Box {stroke [] 0 setdash 2 copy exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V closepath stroke + Pnt} def +/Crs {stroke [] 0 setdash exch hpt sub exch vpt add M + hpt2 vpt2 neg V currentpoint stroke M + hpt2 neg 0 R hpt2 vpt2 V stroke} def +/TriU {stroke [] 0 setdash 2 copy vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V closepath stroke + Pnt} def +/Star {2 copy Pls Crs} def +/BoxF {stroke [] 0 setdash exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V closepath fill} def +/TriUF {stroke [] 0 setdash vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V closepath fill} def +/TriD {stroke [] 0 setdash 2 copy vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V closepath stroke + Pnt} def +/TriDF {stroke [] 0 setdash vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V closepath fill} def +/DiaF {stroke [] 0 setdash vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V closepath fill} def +/Pent {stroke [] 0 setdash 2 copy gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + closepath stroke grestore Pnt} def +/PentF {stroke [] 0 setdash gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + closepath fill grestore} def +/Circle {stroke [] 0 setdash 2 copy + hpt 0 360 arc stroke Pnt} def +/CircleF {stroke [] 0 setdash hpt 0 360 arc fill} def +/C0 {BL [] 0 setdash 2 copy moveto vpt 90 450 arc} bind def +/C1 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 90 arc closepath fill + vpt 0 360 arc closepath} bind def +/C2 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 90 180 arc closepath fill + vpt 0 360 arc closepath} bind def +/C3 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 180 arc closepath fill + vpt 0 360 arc closepath} bind def +/C4 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 180 270 arc closepath fill + vpt 0 360 arc closepath} bind def +/C5 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 90 arc + 2 copy moveto + 2 copy vpt 180 270 arc closepath fill + vpt 0 360 arc} bind def +/C6 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 90 270 arc closepath fill + vpt 0 360 arc closepath} bind def +/C7 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 270 arc closepath fill + vpt 0 360 arc closepath} bind def +/C8 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 270 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C9 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 270 450 arc closepath fill + vpt 0 360 arc closepath} bind def +/C10 {BL [] 0 setdash 2 copy 2 copy moveto vpt 270 360 arc closepath fill + 2 copy moveto + 2 copy vpt 90 180 arc closepath fill + vpt 0 360 arc closepath} bind def +/C11 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 180 arc closepath fill + 2 copy moveto + 2 copy vpt 270 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C12 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 180 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C13 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 90 arc closepath fill + 2 copy moveto + 2 copy vpt 180 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C14 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 90 360 arc closepath fill + vpt 0 360 arc} bind def +/C15 {BL [] 0 setdash 2 copy vpt 0 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/Rec {newpath 4 2 roll moveto 1 index 0 rlineto 0 exch rlineto + neg 0 rlineto closepath} bind def +/Square {dup Rec} bind def +/Bsquare {vpt sub exch vpt sub exch vpt2 Square} bind def +/S0 {BL [] 0 setdash 2 copy moveto 0 vpt rlineto BL Bsquare} bind def +/S1 {BL [] 0 setdash 2 copy vpt Square fill Bsquare} bind def +/S2 {BL [] 0 setdash 2 copy exch vpt sub exch vpt Square fill Bsquare} bind def +/S3 {BL [] 0 setdash 2 copy exch vpt sub exch vpt2 vpt Rec fill Bsquare} bind def +/S4 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt Square fill Bsquare} bind def +/S5 {BL [] 0 setdash 2 copy 2 copy vpt Square fill + exch vpt sub exch vpt sub vpt Square fill Bsquare} bind def +/S6 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt vpt2 Rec fill Bsquare} bind def +/S7 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt vpt2 Rec fill + 2 copy vpt Square fill Bsquare} bind def +/S8 {BL [] 0 setdash 2 copy vpt sub vpt Square fill Bsquare} bind def +/S9 {BL [] 0 setdash 2 copy vpt sub vpt vpt2 Rec fill Bsquare} bind def +/S10 {BL [] 0 setdash 2 copy vpt sub vpt Square fill 2 copy exch vpt sub exch vpt Square fill + Bsquare} bind def +/S11 {BL [] 0 setdash 2 copy vpt sub vpt Square fill 2 copy exch vpt sub exch vpt2 vpt Rec fill + Bsquare} bind def +/S12 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt2 vpt Rec fill Bsquare} bind def +/S13 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt2 vpt Rec fill + 2 copy vpt Square fill Bsquare} bind def +/S14 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt2 vpt Rec fill + 2 copy exch vpt sub exch vpt Square fill Bsquare} bind def +/S15 {BL [] 0 setdash 2 copy Bsquare fill Bsquare} bind def +/D0 {gsave translate 45 rotate 0 0 S0 stroke grestore} bind def +/D1 {gsave translate 45 rotate 0 0 S1 stroke grestore} bind def +/D2 {gsave translate 45 rotate 0 0 S2 stroke grestore} bind def +/D3 {gsave translate 45 rotate 0 0 S3 stroke grestore} bind def +/D4 {gsave translate 45 rotate 0 0 S4 stroke grestore} bind def +/D5 {gsave translate 45 rotate 0 0 S5 stroke grestore} bind def +/D6 {gsave translate 45 rotate 0 0 S6 stroke grestore} bind def +/D7 {gsave translate 45 rotate 0 0 S7 stroke grestore} bind def +/D8 {gsave translate 45 rotate 0 0 S8 stroke grestore} bind def +/D9 {gsave translate 45 rotate 0 0 S9 stroke grestore} bind def +/D10 {gsave translate 45 rotate 0 0 S10 stroke grestore} bind def +/D11 {gsave translate 45 rotate 0 0 S11 stroke grestore} bind def +/D12 {gsave translate 45 rotate 0 0 S12 stroke grestore} bind def +/D13 {gsave translate 45 rotate 0 0 S13 stroke grestore} bind def +/D14 {gsave translate 45 rotate 0 0 S14 stroke grestore} bind def +/D15 {gsave translate 45 rotate 0 0 S15 stroke grestore} bind def +/DiaE {stroke [] 0 setdash vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V closepath stroke} def +/BoxE {stroke [] 0 setdash exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V closepath stroke} def +/TriUE {stroke [] 0 setdash vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V closepath stroke} def +/TriDE {stroke [] 0 setdash vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V closepath stroke} def +/PentE {stroke [] 0 setdash gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + closepath stroke grestore} def +/CircE {stroke [] 0 setdash + hpt 0 360 arc stroke} def +/Opaque {gsave closepath 1 setgray fill grestore 0 setgray closepath} def +/DiaW {stroke [] 0 setdash vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V Opaque stroke} def +/BoxW {stroke [] 0 setdash exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V Opaque stroke} def +/TriUW {stroke [] 0 setdash vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V Opaque stroke} def +/TriDW {stroke [] 0 setdash vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V Opaque stroke} def +/PentW {stroke [] 0 setdash gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + Opaque stroke grestore} def +/CircW {stroke [] 0 setdash + hpt 0 360 arc Opaque stroke} def +/BoxFill {gsave Rec 1 setgray fill grestore} def +/Density { + /Fillden exch def + currentrgbcolor + /ColB exch def /ColG exch def /ColR exch def + /ColR ColR Fillden mul Fillden sub 1 add def + /ColG ColG Fillden mul Fillden sub 1 add def + /ColB ColB Fillden mul Fillden sub 1 add def + ColR ColG ColB setrgbcolor} def +/BoxColFill {gsave Rec PolyFill} def +/PolyFill {gsave Density fill grestore grestore} def +/h {rlineto rlineto rlineto gsave closepath fill grestore} bind def +% +% PostScript Level 1 Pattern Fill routine for rectangles +% Usage: x y w h s a XX PatternFill +% x,y = lower left corner of box to be filled +% w,h = width and height of box +% a = angle in degrees between lines and x-axis +% XX = 0/1 for no/yes cross-hatch +% +/PatternFill {gsave /PFa [ 9 2 roll ] def + PFa 0 get PFa 2 get 2 div add PFa 1 get PFa 3 get 2 div add translate + PFa 2 get -2 div PFa 3 get -2 div PFa 2 get PFa 3 get Rec + gsave 1 setgray fill grestore clip + currentlinewidth 0.5 mul setlinewidth + /PFs PFa 2 get dup mul PFa 3 get dup mul add sqrt def + 0 0 M PFa 5 get rotate PFs -2 div dup translate + 0 1 PFs PFa 4 get div 1 add floor cvi + {PFa 4 get mul 0 M 0 PFs V} for + 0 PFa 6 get ne { + 0 1 PFs PFa 4 get div 1 add floor cvi + {PFa 4 get mul 0 2 1 roll M PFs 0 V} for + } if + stroke grestore} def +% +/languagelevel where + {pop languagelevel} {1} ifelse + 2 lt + {/InterpretLevel1 true def} + {/InterpretLevel1 Level1 def} + ifelse +% +% PostScript level 2 pattern fill definitions +% +/Level2PatternFill { +/Tile8x8 {/PaintType 2 /PatternType 1 /TilingType 1 /BBox [0 0 8 8] /XStep 8 /YStep 8} + bind def +/KeepColor {currentrgbcolor [/Pattern /DeviceRGB] setcolorspace} bind def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 0 M 8 8 L 0 8 M 8 0 L stroke} +>> matrix makepattern +/Pat1 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 0 M 8 8 L 0 8 M 8 0 L stroke + 0 4 M 4 8 L 8 4 L 4 0 L 0 4 L stroke} +>> matrix makepattern +/Pat2 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 0 M 0 8 L + 8 8 L 8 0 L 0 0 L fill} +>> matrix makepattern +/Pat3 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -4 8 M 8 -4 L + 0 12 M 12 0 L stroke} +>> matrix makepattern +/Pat4 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -4 0 M 8 12 L + 0 -4 M 12 8 L stroke} +>> matrix makepattern +/Pat5 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -2 8 M 4 -4 L + 0 12 M 8 -4 L 4 12 M 10 0 L stroke} +>> matrix makepattern +/Pat6 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -2 0 M 4 12 L + 0 -4 M 8 12 L 4 -4 M 10 8 L stroke} +>> matrix makepattern +/Pat7 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 8 -2 M -4 4 L + 12 0 M -4 8 L 12 4 M 0 10 L stroke} +>> matrix makepattern +/Pat8 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 -2 M 12 4 L + -4 0 M 12 8 L -4 4 M 8 10 L stroke} +>> matrix makepattern +/Pat9 exch def +/Pattern1 {PatternBgnd KeepColor Pat1 setpattern} bind def +/Pattern2 {PatternBgnd KeepColor Pat2 setpattern} bind def +/Pattern3 {PatternBgnd KeepColor Pat3 setpattern} bind def +/Pattern4 {PatternBgnd KeepColor Landscape {Pat5} {Pat4} ifelse setpattern} bind def +/Pattern5 {PatternBgnd KeepColor Landscape {Pat4} {Pat5} ifelse setpattern} bind def +/Pattern6 {PatternBgnd KeepColor Landscape {Pat9} {Pat6} ifelse setpattern} bind def +/Pattern7 {PatternBgnd KeepColor Landscape {Pat8} {Pat7} ifelse setpattern} bind def +} def +% +% +%End of PostScript Level 2 code +% +/PatternBgnd { + TransparentPatterns {} {gsave 1 setgray fill grestore} ifelse +} def +% +% Substitute for Level 2 pattern fill codes with +% grayscale if Level 2 support is not selected. +% +/Level1PatternFill { +/Pattern1 {0.250 Density} bind def +/Pattern2 {0.500 Density} bind def +/Pattern3 {0.750 Density} bind def +/Pattern4 {0.125 Density} bind def +/Pattern5 {0.375 Density} bind def +/Pattern6 {0.625 Density} bind def +/Pattern7 {0.875 Density} bind def +} def +% +% Now test for support of Level 2 code +% +Level1 {Level1PatternFill} {Level2PatternFill} ifelse +% +/Symbol-Oblique /Symbol findfont [1 0 .167 1 0 0] makefont +dup length dict begin {1 index /FID eq {pop pop} {def} ifelse} forall +currentdict end definefont pop +Level1 SuppressPDFMark or +{} { +/SDict 10 dict def +systemdict /pdfmark known not { + userdict /pdfmark systemdict /cleartomark get put +} if +SDict begin [ + /Title (energy.eps) + /Subject (gnuplot plot) + /Creator (gnuplot 4.6 patchlevel 0) + /Author (afanfakh) +% /Producer (gnuplot) +% /Keywords () + /CreationDate (Thu Feb 19 16:44:03 2015) + /DOCINFO pdfmark +end +} ifelse +end +%%EndProlog +%%Page: 1 1 +gnudict begin +gsave +doclip +50 50 translate +0.050 0.050 scale +0 setgray +newpath +(Helvetica) findfont 140 scalefont setfont +BackgroundColor 0 lt 3 1 roll 0 lt exch 0 lt or or not {BackgroundColor C 1.000 0 0 5400.00 3780.00 BoxColFill} if +1.000 UL +LTb +602 588 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +518 588 M +( 0) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 948 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +518 948 M +( 5) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 1308 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 10) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 1668 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 15) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 2028 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 20) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 2387 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 25) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 2747 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 30) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 3107 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 35) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 3467 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 40) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +604 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +604 448 M +( 0) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +1088 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 16) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +1572 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 32) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +2057 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 48) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +2541 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 64) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +3026 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 80) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +3510 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 96) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +3995 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 112) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +4479 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 128) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +4964 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 144) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +1.000 UL +LTb +602 3611 N +602 588 L +4545 0 V +0 3023 V +-4545 0 V +Z stroke +LCb setrgbcolor +/Helvetica findfont 210 scalefont setfont +112 2099 M +currentpoint gsave translate -270 rotate 0 0 M +(Energy saving) Cshow +grestore +/Helvetica findfont 140 scalefont setfont +LTb +LCb setrgbcolor +/Helvetica findfont 210 scalefont setfont +2974 98 M +( Number of nodes) Cshow +/Helvetica findfont 140 scalefont setfont +LTb +1.000 UP +/Helvetica findfont 190 scalefont setfont +649 620 M +( ) Lshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +% Begin plot #1 +1.500 UP +2.000 UL +LT0 +0.00 0.00 1.00 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +1259 3443 M +(CG) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.00 0.00 1.00 C 709 3443 M +298 0 V +725 3047 M +846 2838 L +242 -357 V +484 -730 V +969 -578 V +4479 876 L +725 3047 Box +846 2838 Box +1088 2481 Box +1572 1751 Box +2541 1173 Box +4479 876 Box +858 3443 Box +% End plot #1 +% Begin plot #2 +1.500 UP +2.000 UL +LT0 +1.00 0.00 0.00 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +1893 3443 M +(MG) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +1.00 0.00 0.00 C 1343 3443 M +298 0 V +725 3134 M +846 3020 L +242 -93 V +484 -256 V +969 24 V +4479 1940 L +725 3134 TriD +846 3020 TriD +1088 2927 TriD +1572 2671 TriD +2541 2695 TriD +4479 1940 TriD +1492 3443 TriD +% End plot #2 +% Begin plot #3 +1.500 UP +2.000 UL +LT0 +0.50 0.00 0.50 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +2527 3443 M +(EP) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.50 0.00 0.50 C 1977 3443 M +298 0 V +725 2519 M +121 15 V +242 -13 V +484 9 V +969 10 V +1938 -2 V +725 2519 Star +846 2534 Star +1088 2521 Star +1572 2530 Star +2541 2540 Star +4479 2538 Star +2126 3443 Star +% End plot #3 +% Begin plot #4 +1.500 UP +2.000 UL +LT0 +0.18 0.31 0.31 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +3161 3443 M +(LU) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.18 0.31 0.31 C 2611 3443 M +298 0 V +725 3036 M +846 2620 L +242 41 V +484 -374 V +969 -118 V +4479 1788 L +725 3036 TriUF +846 2620 TriUF +1088 2661 TriUF +1572 2287 TriUF +2541 2169 TriUF +4479 1788 TriUF +2760 3443 TriUF +% End plot #4 +% Begin plot #5 +1.500 UP +2.000 UL +LT0 +0.18 0.55 0.34 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +3795 3443 M +(BT) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.18 0.55 0.34 C 3245 3443 M +298 0 V +725 3133 M +876 2915 L +212 189 V +606 -24 V +847 -187 V +4964 2257 L +725 3133 BoxF +876 2915 BoxF +1088 3104 BoxF +1694 3080 BoxF +2541 2893 BoxF +4964 2257 BoxF +3394 3443 BoxF +% End plot #5 +% Begin plot #6 +1.500 UP +2.000 UL +LT0 +0.85 0.65 0.13 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +4429 3443 M +(SP) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.85 0.65 0.13 C 3879 3443 M +298 0 V +725 3123 M +876 2368 L +212 -161 V +606 17 V +847 152 V +2423 136 V +725 3123 Circle +876 2368 Circle +1088 2207 Circle +1694 2224 Circle +2541 2376 Circle +4964 2512 Circle +4028 3443 Circle +% End plot #6 +% Begin plot #7 +1.500 UP +2.000 UL +LT0 +0.55 0.00 0.00 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +5063 3443 M +(FT) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.55 0.00 0.00 C 4513 3443 M +298 0 V +725 3148 M +846 2821 L +242 41 V +484 -612 V +969 -212 V +4479 1513 L +725 3148 CircleF +846 2821 CircleF +1088 2862 CircleF +1572 2250 CircleF +2541 2038 CircleF +4479 1513 CircleF +4662 3443 CircleF +% End plot #7 +1.000 UL +LTb +602 3611 N +602 588 L +4545 0 V +0 3023 V +-4545 0 V +Z stroke +1.000 UP +1.000 UL +LTb +stroke +grestore +end +showpage diff --git a/fig/heter.pdf b/fig/heter-eps-converted-to.pdf similarity index 93% rename from fig/heter.pdf rename to fig/heter-eps-converted-to.pdf index 006ad27..34a7fad 100644 Binary files a/fig/heter.pdf and b/fig/heter-eps-converted-to.pdf differ diff --git a/fig/heter.eps b/fig/heter.eps index e012be1..b1b59f5 100644 --- a/fig/heter.eps +++ b/fig/heter.eps @@ -1,7 +1,7 @@ %!PS-Adobe-2.0 EPSF-2.0 %%Title: heter2.eps %%Creator: gnuplot 4.6 patchlevel 0 -%%CreationDate: Fri May 15 14:32:08 2015 +%%CreationDate: Thu Feb 19 12:00:04 2015 %%DocumentFonts: (atend) %%BoundingBox: 50 50 320 239 %%EndComments @@ -432,7 +432,7 @@ SDict begin [ /Author (afanfakh) % /Producer (gnuplot) % /Keywords () - /CreationDate (Fri May 15 14:32:08 2015) + /CreationDate (Thu Feb 19 12:00:04 2015) /DOCINFO pdfmark end } ifelse @@ -524,77 +524,55 @@ LCb setrgbcolor LTb 1.000 UP /Helvetica findfont 220 scalefont setfont -1435 558 M +1801 558 M (Optimal vector of scaling factors) Lshow /Helvetica findfont 140 scalefont setfont /Helvetica findfont 220 scalefont setfont -4255 1497 M +5147 1497 M () Lshow /Helvetica findfont 140 scalefont setfont /Helvetica findfont 190 scalefont setfont -793 3277 M +820 3277 M ( Upper bound) Lshow /Helvetica findfont 140 scalefont setfont -/Helvetica findfont 190 scalefont setfont -3541 1887 M -( Lower bound) Lshow -/Helvetica findfont 140 scalefont setfont -1384 2246 M -( Max.) Lshow -1384 2102 M -( distance) Lshow 1.000 UL LTb -1378 2821 M -0 -1153 V +1734 2688 M +0 -1265 V stroke LT2 -1346 517 M +1702 517 M 32 -121 V 32 121 V --32 2304 R -0 -2425 V +-32 2171 R +0 -2292 V stroke gsave [] 0 setdash -1346 517 M +1702 517 M 32 -121 V 32 121 V stroke grestore 1.000 UL LT2 -1217 2976 M +1355 2976 M 22 -83 V 22 83 V -22 204 R 0 -287 V stroke gsave [] 0 setdash -1217 2976 M +1355 2976 M 22 -83 V 22 83 V stroke grestore 1.000 UL -LT2 -4011 1534 M -30 -114 V -30 114 V --30 281 R -0 -395 V -stroke -gsave [] 0 setdash -4011 1534 M -30 -114 V -30 114 V -stroke -grestore -1.000 UL LT0 -1311 2821 M -142 0 V -1311 1668 M -142 0 V +1623 2688 M +234 0 V +1623 1423 M +234 0 V stroke LTb % Begin plot #1 @@ -611,24 +589,22 @@ LC2 setrgbcolor 4580 3443 M 399 0 V 686 2893 M -74 0 V -81 0 V -88 0 V -97 0 V -106 0 V -117 0 V -131 -72 V -146 -142 V -164 -135 V -186 -171 V -212 -164 V -245 -148 V -286 -164 V -338 -170 V -406 -135 V -495 -135 V -620 -135 V -312 -63 V +93 0 V +101 0 V +110 0 V +121 0 V +133 0 V +146 0 V +163 -32 V +183 -173 V +205 -178 V +232 -183 V +266 -190 V +306 -195 V +357 -201 V +423 -129 V +507 -79 V +619 -36 V % End plot #1 % Begin plot #2 stroke @@ -637,30 +613,28 @@ LT0 1.00 0.00 0.00 C LCb setrgbcolor /Helvetica findfont 220 scalefont setfont 4496 3233 M -(Normalized energy) Rshow +(Normalize energy) Rshow /Helvetica findfont 140 scalefont setfont LT0 1.00 0.00 0.00 C 4580 3233 M 399 0 V 686 2893 M -74 -248 V -81 -193 V -88 -182 V -97 -170 V -106 -156 V -117 -144 V -131 -129 V -146 -115 V -164 -100 V -186 -82 V -212 -64 V -245 -43 V -286 -20 V -338 7 V -406 39 V -495 80 V -620 131 V -312 59 V +93 -144 V +880 2494 L +990 2292 L +121 -159 V +133 -203 V +146 -177 V +163 -151 V +183 -177 V +205 -129 V +232 -89 V +266 -67 V +306 -43 V +357 -36 V +423 -33 V +507 -28 V +619 0 V % End plot #2 % Begin plot #3 stroke @@ -683,7 +657,3 @@ stroke grestore end showpage -%%Trailer -%%DocumentFonts: Helvetica -%%Trailer -%%DocumentFonts: Helvetica diff --git a/fig/per_deg.pdf b/fig/per_deg-eps-converted-to.pdf similarity index 95% rename from fig/per_deg.pdf rename to fig/per_deg-eps-converted-to.pdf index e4211b5..821f861 100644 Binary files a/fig/per_deg.pdf and b/fig/per_deg-eps-converted-to.pdf differ diff --git a/fig/per_deg.eps b/fig/per_deg.eps new file mode 100644 index 0000000..e1ef706 --- /dev/null +++ b/fig/per_deg.eps @@ -0,0 +1,820 @@ +%!PS-Adobe-2.0 EPSF-2.0 +%%Title: per_deg.eps +%%Creator: gnuplot 4.6 patchlevel 0 +%%CreationDate: Thu Feb 19 16:42:46 2015 +%%DocumentFonts: (atend) +%%BoundingBox: 50 50 320 239 +%%EndComments +%%BeginProlog +/gnudict 256 dict def +gnudict begin +% +% The following true/false flags may be edited by hand if desired. +% The unit line width and grayscale image gamma correction may also be changed. +% +/Color false def +/Blacktext false def +/Solid false def +/Dashlength 1 def +/Landscape false def +/Level1 false def +/Rounded false def +/ClipToBoundingBox false def +/SuppressPDFMark false def +/TransparentPatterns false def +/gnulinewidth 5.000 def +/userlinewidth gnulinewidth def +/Gamma 1.0 def +/BackgroundColor {-1.000 -1.000 -1.000} def +% +/vshift -46 def +/dl1 { + 10.0 Dashlength mul mul + Rounded { currentlinewidth 0.75 mul sub dup 0 le { pop 0.01 } if } if +} def +/dl2 { + 10.0 Dashlength mul mul + Rounded { currentlinewidth 0.75 mul add } if +} def +/hpt_ 31.5 def +/vpt_ 31.5 def +/hpt hpt_ def +/vpt vpt_ def +/doclip { + ClipToBoundingBox { + newpath 50 50 moveto 320 50 lineto 320 239 lineto 50 239 lineto closepath + clip + } if +} def +% +% Gnuplot Prolog Version 4.4 (August 2010) +% +%/SuppressPDFMark true def +% +/M {moveto} bind def +/L {lineto} bind def +/R {rmoveto} bind def +/V {rlineto} bind def +/N {newpath moveto} bind def +/Z {closepath} bind def +/C {setrgbcolor} bind def +/f {rlineto fill} bind def +/g {setgray} bind def +/Gshow {show} def % May be redefined later in the file to support UTF-8 +/vpt2 vpt 2 mul def +/hpt2 hpt 2 mul def +/Lshow {currentpoint stroke M 0 vshift R + Blacktext {gsave 0 setgray show grestore} {show} ifelse} def +/Rshow {currentpoint stroke M dup stringwidth pop neg vshift R + Blacktext {gsave 0 setgray show grestore} {show} ifelse} def +/Cshow {currentpoint stroke M dup stringwidth pop -2 div vshift R + Blacktext {gsave 0 setgray show grestore} {show} ifelse} def +/UP {dup vpt_ mul /vpt exch def hpt_ mul /hpt exch def + /hpt2 hpt 2 mul def /vpt2 vpt 2 mul def} def +/DL {Color {setrgbcolor Solid {pop []} if 0 setdash} + {pop pop pop 0 setgray Solid {pop []} if 0 setdash} ifelse} def +/BL {stroke userlinewidth 2 mul setlinewidth + Rounded {1 setlinejoin 1 setlinecap} if} def +/AL {stroke userlinewidth 2 div setlinewidth + Rounded {1 setlinejoin 1 setlinecap} if} def +/UL {dup gnulinewidth mul /userlinewidth exch def + dup 1 lt {pop 1} if 10 mul /udl exch def} def +/PL {stroke userlinewidth setlinewidth + Rounded {1 setlinejoin 1 setlinecap} if} def +3.8 setmiterlimit +% Default Line colors +/LCw {1 1 1} def +/LCb {0 0 0} def +/LCa {0 0 0} def +/LC0 {1 0 0} def +/LC1 {0 1 0} def +/LC2 {0 0 1} def +/LC3 {1 0 1} def +/LC4 {0 1 1} def +/LC5 {1 1 0} def +/LC6 {0 0 0} def +/LC7 {1 0.3 0} def +/LC8 {0.5 0.5 0.5} def +% Default Line Types +/LTw {PL [] 1 setgray} def +/LTb {BL [] LCb DL} def +/LTa {AL [1 udl mul 2 udl mul] 0 setdash LCa setrgbcolor} def +/LT0 {PL [] LC0 DL} def +/LT1 {PL [4 dl1 2 dl2] LC1 DL} def +/LT2 {PL [2 dl1 3 dl2] LC2 DL} def +/LT3 {PL [1 dl1 1.5 dl2] LC3 DL} def +/LT4 {PL [6 dl1 2 dl2 1 dl1 2 dl2] LC4 DL} def +/LT5 {PL [3 dl1 3 dl2 1 dl1 3 dl2] LC5 DL} def +/LT6 {PL [2 dl1 2 dl2 2 dl1 6 dl2] LC6 DL} def +/LT7 {PL [1 dl1 2 dl2 6 dl1 2 dl2 1 dl1 2 dl2] LC7 DL} def +/LT8 {PL [2 dl1 2 dl2 2 dl1 2 dl2 2 dl1 2 dl2 2 dl1 4 dl2] LC8 DL} def +/Pnt {stroke [] 0 setdash gsave 1 setlinecap M 0 0 V stroke grestore} def +/Dia {stroke [] 0 setdash 2 copy vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V closepath stroke + Pnt} def +/Pls {stroke [] 0 setdash vpt sub M 0 vpt2 V + currentpoint stroke M + hpt neg vpt neg R hpt2 0 V stroke + } def +/Box {stroke [] 0 setdash 2 copy exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V closepath stroke + Pnt} def +/Crs {stroke [] 0 setdash exch hpt sub exch vpt add M + hpt2 vpt2 neg V currentpoint stroke M + hpt2 neg 0 R hpt2 vpt2 V stroke} def +/TriU {stroke [] 0 setdash 2 copy vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V closepath stroke + Pnt} def +/Star {2 copy Pls Crs} def +/BoxF {stroke [] 0 setdash exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V closepath fill} def +/TriUF {stroke [] 0 setdash vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V closepath fill} def +/TriD {stroke [] 0 setdash 2 copy vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V closepath stroke + Pnt} def +/TriDF {stroke [] 0 setdash vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V closepath fill} def +/DiaF {stroke [] 0 setdash vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V closepath fill} def +/Pent {stroke [] 0 setdash 2 copy gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + closepath stroke grestore Pnt} def +/PentF {stroke [] 0 setdash gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + closepath fill grestore} def +/Circle {stroke [] 0 setdash 2 copy + hpt 0 360 arc stroke Pnt} def +/CircleF {stroke [] 0 setdash hpt 0 360 arc fill} def +/C0 {BL [] 0 setdash 2 copy moveto vpt 90 450 arc} bind def +/C1 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 90 arc closepath fill + vpt 0 360 arc closepath} bind def +/C2 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 90 180 arc closepath fill + vpt 0 360 arc closepath} bind def +/C3 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 180 arc closepath fill + vpt 0 360 arc closepath} bind def +/C4 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 180 270 arc closepath fill + vpt 0 360 arc closepath} bind def +/C5 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 90 arc + 2 copy moveto + 2 copy vpt 180 270 arc closepath fill + vpt 0 360 arc} bind def +/C6 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 90 270 arc closepath fill + vpt 0 360 arc closepath} bind def +/C7 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 270 arc closepath fill + vpt 0 360 arc closepath} bind def +/C8 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 270 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C9 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 270 450 arc closepath fill + vpt 0 360 arc closepath} bind def +/C10 {BL [] 0 setdash 2 copy 2 copy moveto vpt 270 360 arc closepath fill + 2 copy moveto + 2 copy vpt 90 180 arc closepath fill + vpt 0 360 arc closepath} bind def +/C11 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 180 arc closepath fill + 2 copy moveto + 2 copy vpt 270 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C12 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 180 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C13 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 0 90 arc closepath fill + 2 copy moveto + 2 copy vpt 180 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/C14 {BL [] 0 setdash 2 copy moveto + 2 copy vpt 90 360 arc closepath fill + vpt 0 360 arc} bind def +/C15 {BL [] 0 setdash 2 copy vpt 0 360 arc closepath fill + vpt 0 360 arc closepath} bind def +/Rec {newpath 4 2 roll moveto 1 index 0 rlineto 0 exch rlineto + neg 0 rlineto closepath} bind def +/Square {dup Rec} bind def +/Bsquare {vpt sub exch vpt sub exch vpt2 Square} bind def +/S0 {BL [] 0 setdash 2 copy moveto 0 vpt rlineto BL Bsquare} bind def +/S1 {BL [] 0 setdash 2 copy vpt Square fill Bsquare} bind def +/S2 {BL [] 0 setdash 2 copy exch vpt sub exch vpt Square fill Bsquare} bind def +/S3 {BL [] 0 setdash 2 copy exch vpt sub exch vpt2 vpt Rec fill Bsquare} bind def +/S4 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt Square fill Bsquare} bind def +/S5 {BL [] 0 setdash 2 copy 2 copy vpt Square fill + exch vpt sub exch vpt sub vpt Square fill Bsquare} bind def +/S6 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt vpt2 Rec fill Bsquare} bind def +/S7 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt vpt2 Rec fill + 2 copy vpt Square fill Bsquare} bind def +/S8 {BL [] 0 setdash 2 copy vpt sub vpt Square fill Bsquare} bind def +/S9 {BL [] 0 setdash 2 copy vpt sub vpt vpt2 Rec fill Bsquare} bind def +/S10 {BL [] 0 setdash 2 copy vpt sub vpt Square fill 2 copy exch vpt sub exch vpt Square fill + Bsquare} bind def +/S11 {BL [] 0 setdash 2 copy vpt sub vpt Square fill 2 copy exch vpt sub exch vpt2 vpt Rec fill + Bsquare} bind def +/S12 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt2 vpt Rec fill Bsquare} bind def +/S13 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt2 vpt Rec fill + 2 copy vpt Square fill Bsquare} bind def +/S14 {BL [] 0 setdash 2 copy exch vpt sub exch vpt sub vpt2 vpt Rec fill + 2 copy exch vpt sub exch vpt Square fill Bsquare} bind def +/S15 {BL [] 0 setdash 2 copy Bsquare fill Bsquare} bind def +/D0 {gsave translate 45 rotate 0 0 S0 stroke grestore} bind def +/D1 {gsave translate 45 rotate 0 0 S1 stroke grestore} bind def +/D2 {gsave translate 45 rotate 0 0 S2 stroke grestore} bind def +/D3 {gsave translate 45 rotate 0 0 S3 stroke grestore} bind def +/D4 {gsave translate 45 rotate 0 0 S4 stroke grestore} bind def +/D5 {gsave translate 45 rotate 0 0 S5 stroke grestore} bind def +/D6 {gsave translate 45 rotate 0 0 S6 stroke grestore} bind def +/D7 {gsave translate 45 rotate 0 0 S7 stroke grestore} bind def +/D8 {gsave translate 45 rotate 0 0 S8 stroke grestore} bind def +/D9 {gsave translate 45 rotate 0 0 S9 stroke grestore} bind def +/D10 {gsave translate 45 rotate 0 0 S10 stroke grestore} bind def +/D11 {gsave translate 45 rotate 0 0 S11 stroke grestore} bind def +/D12 {gsave translate 45 rotate 0 0 S12 stroke grestore} bind def +/D13 {gsave translate 45 rotate 0 0 S13 stroke grestore} bind def +/D14 {gsave translate 45 rotate 0 0 S14 stroke grestore} bind def +/D15 {gsave translate 45 rotate 0 0 S15 stroke grestore} bind def +/DiaE {stroke [] 0 setdash vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V closepath stroke} def +/BoxE {stroke [] 0 setdash exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V closepath stroke} def +/TriUE {stroke [] 0 setdash vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V closepath stroke} def +/TriDE {stroke [] 0 setdash vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V closepath stroke} def +/PentE {stroke [] 0 setdash gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + closepath stroke grestore} def +/CircE {stroke [] 0 setdash + hpt 0 360 arc stroke} def +/Opaque {gsave closepath 1 setgray fill grestore 0 setgray closepath} def +/DiaW {stroke [] 0 setdash vpt add M + hpt neg vpt neg V hpt vpt neg V + hpt vpt V hpt neg vpt V Opaque stroke} def +/BoxW {stroke [] 0 setdash exch hpt sub exch vpt add M + 0 vpt2 neg V hpt2 0 V 0 vpt2 V + hpt2 neg 0 V Opaque stroke} def +/TriUW {stroke [] 0 setdash vpt 1.12 mul add M + hpt neg vpt -1.62 mul V + hpt 2 mul 0 V + hpt neg vpt 1.62 mul V Opaque stroke} def +/TriDW {stroke [] 0 setdash vpt 1.12 mul sub M + hpt neg vpt 1.62 mul V + hpt 2 mul 0 V + hpt neg vpt -1.62 mul V Opaque stroke} def +/PentW {stroke [] 0 setdash gsave + translate 0 hpt M 4 {72 rotate 0 hpt L} repeat + Opaque stroke grestore} def +/CircW {stroke [] 0 setdash + hpt 0 360 arc Opaque stroke} def +/BoxFill {gsave Rec 1 setgray fill grestore} def +/Density { + /Fillden exch def + currentrgbcolor + /ColB exch def /ColG exch def /ColR exch def + /ColR ColR Fillden mul Fillden sub 1 add def + /ColG ColG Fillden mul Fillden sub 1 add def + /ColB ColB Fillden mul Fillden sub 1 add def + ColR ColG ColB setrgbcolor} def +/BoxColFill {gsave Rec PolyFill} def +/PolyFill {gsave Density fill grestore grestore} def +/h {rlineto rlineto rlineto gsave closepath fill grestore} bind def +% +% PostScript Level 1 Pattern Fill routine for rectangles +% Usage: x y w h s a XX PatternFill +% x,y = lower left corner of box to be filled +% w,h = width and height of box +% a = angle in degrees between lines and x-axis +% XX = 0/1 for no/yes cross-hatch +% +/PatternFill {gsave /PFa [ 9 2 roll ] def + PFa 0 get PFa 2 get 2 div add PFa 1 get PFa 3 get 2 div add translate + PFa 2 get -2 div PFa 3 get -2 div PFa 2 get PFa 3 get Rec + gsave 1 setgray fill grestore clip + currentlinewidth 0.5 mul setlinewidth + /PFs PFa 2 get dup mul PFa 3 get dup mul add sqrt def + 0 0 M PFa 5 get rotate PFs -2 div dup translate + 0 1 PFs PFa 4 get div 1 add floor cvi + {PFa 4 get mul 0 M 0 PFs V} for + 0 PFa 6 get ne { + 0 1 PFs PFa 4 get div 1 add floor cvi + {PFa 4 get mul 0 2 1 roll M PFs 0 V} for + } if + stroke grestore} def +% +/languagelevel where + {pop languagelevel} {1} ifelse + 2 lt + {/InterpretLevel1 true def} + {/InterpretLevel1 Level1 def} + ifelse +% +% PostScript level 2 pattern fill definitions +% +/Level2PatternFill { +/Tile8x8 {/PaintType 2 /PatternType 1 /TilingType 1 /BBox [0 0 8 8] /XStep 8 /YStep 8} + bind def +/KeepColor {currentrgbcolor [/Pattern /DeviceRGB] setcolorspace} bind def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 0 M 8 8 L 0 8 M 8 0 L stroke} +>> matrix makepattern +/Pat1 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 0 M 8 8 L 0 8 M 8 0 L stroke + 0 4 M 4 8 L 8 4 L 4 0 L 0 4 L stroke} +>> matrix makepattern +/Pat2 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 0 M 0 8 L + 8 8 L 8 0 L 0 0 L fill} +>> matrix makepattern +/Pat3 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -4 8 M 8 -4 L + 0 12 M 12 0 L stroke} +>> matrix makepattern +/Pat4 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -4 0 M 8 12 L + 0 -4 M 12 8 L stroke} +>> matrix makepattern +/Pat5 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -2 8 M 4 -4 L + 0 12 M 8 -4 L 4 12 M 10 0 L stroke} +>> matrix makepattern +/Pat6 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop -2 0 M 4 12 L + 0 -4 M 8 12 L 4 -4 M 10 8 L stroke} +>> matrix makepattern +/Pat7 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 8 -2 M -4 4 L + 12 0 M -4 8 L 12 4 M 0 10 L stroke} +>> matrix makepattern +/Pat8 exch def +<< Tile8x8 + /PaintProc {0.5 setlinewidth pop 0 -2 M 12 4 L + -4 0 M 12 8 L -4 4 M 8 10 L stroke} +>> matrix makepattern +/Pat9 exch def +/Pattern1 {PatternBgnd KeepColor Pat1 setpattern} bind def +/Pattern2 {PatternBgnd KeepColor Pat2 setpattern} bind def +/Pattern3 {PatternBgnd KeepColor Pat3 setpattern} bind def +/Pattern4 {PatternBgnd KeepColor Landscape {Pat5} {Pat4} ifelse setpattern} bind def +/Pattern5 {PatternBgnd KeepColor Landscape {Pat4} {Pat5} ifelse setpattern} bind def +/Pattern6 {PatternBgnd KeepColor Landscape {Pat9} {Pat6} ifelse setpattern} bind def +/Pattern7 {PatternBgnd KeepColor Landscape {Pat8} {Pat7} ifelse setpattern} bind def +} def +% +% +%End of PostScript Level 2 code +% +/PatternBgnd { + TransparentPatterns {} {gsave 1 setgray fill grestore} ifelse +} def +% +% Substitute for Level 2 pattern fill codes with +% grayscale if Level 2 support is not selected. +% +/Level1PatternFill { +/Pattern1 {0.250 Density} bind def +/Pattern2 {0.500 Density} bind def +/Pattern3 {0.750 Density} bind def +/Pattern4 {0.125 Density} bind def +/Pattern5 {0.375 Density} bind def +/Pattern6 {0.625 Density} bind def +/Pattern7 {0.875 Density} bind def +} def +% +% Now test for support of Level 2 code +% +Level1 {Level1PatternFill} {Level2PatternFill} ifelse +% +/Symbol-Oblique /Symbol findfont [1 0 .167 1 0 0] makefont +dup length dict begin {1 index /FID eq {pop pop} {def} ifelse} forall +currentdict end definefont pop +Level1 SuppressPDFMark or +{} { +/SDict 10 dict def +systemdict /pdfmark known not { + userdict /pdfmark systemdict /cleartomark get put +} if +SDict begin [ + /Title (per_deg.eps) + /Subject (gnuplot plot) + /Creator (gnuplot 4.6 patchlevel 0) + /Author (afanfakh) +% /Producer (gnuplot) +% /Keywords () + /CreationDate (Thu Feb 19 16:42:46 2015) + /DOCINFO pdfmark +end +} ifelse +end +%%EndProlog +%%Page: 1 1 +gnudict begin +gsave +doclip +50 50 translate +0.050 0.050 scale +0 setgray +newpath +(Helvetica) findfont 140 scalefont setfont +BackgroundColor 0 lt 3 1 roll 0 lt exch 0 lt or or not {BackgroundColor C 1.000 0 0 5400.00 3780.00 BoxColFill} if +1.000 UL +LTb +602 674 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +518 674 M +( 0) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 1538 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 5) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 2402 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 10) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +602 3266 M +63 0 V +4482 0 R +-63 0 V +/Helvetica findfont 190 scalefont setfont +-4566 0 R +( 15) Rshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +604 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +604 448 M +( 0) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +1088 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 16) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +1572 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 32) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +2057 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 48) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +2541 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 64) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +3026 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 80) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +3510 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 96) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +3995 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 112) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +4479 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 128) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +4964 588 M +0 63 V +0 2960 R +0 -63 V +/Helvetica findfont 190 scalefont setfont +0 -3100 R +( 144) Cshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +1.000 UL +LTb +602 3611 N +602 588 L +4545 0 V +0 3023 V +-4545 0 V +Z stroke +LCb setrgbcolor +/Helvetica findfont 210 scalefont setfont +112 2099 M +currentpoint gsave translate -270 rotate 0 0 M +(Performance degradation) Cshow +grestore +/Helvetica findfont 140 scalefont setfont +LTb +LCb setrgbcolor +/Helvetica findfont 210 scalefont setfont +2974 98 M +( Number of nodes) Cshow +/Helvetica findfont 140 scalefont setfont +LTb +1.000 UP +/Helvetica findfont 190 scalefont setfont +649 752 M +( ) Lshow +/Helvetica findfont 140 scalefont setfont +1.000 UL +LTb +% Begin plot #1 +1.500 UP +2.000 UL +LT0 +0.00 0.00 1.00 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +1259 3443 M +(CG) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.00 0.00 1.00 C 709 3443 M +298 0 V +725 1835 M +121 70 V +242 423 V +484 -737 V +2541 965 L +4479 865 L +725 1835 Box +846 1905 Box +1088 2328 Box +1572 1591 Box +2541 965 Box +4479 865 Box +858 3443 Box +% End plot #1 +% Begin plot #2 +1.500 UP +2.000 UL +LT0 +1.00 0.00 0.00 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +1893 3443 M +(MG) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +1.00 0.00 0.00 C 1343 3443 M +298 0 V +725 1424 M +121 359 V +242 -63 V +484 483 V +969 951 V +4479 2469 L +725 1424 TriD +846 1783 TriD +1088 1720 TriD +1572 2203 TriD +2541 3154 TriD +4479 2469 TriD +1492 3443 TriD +% End plot #2 +% Begin plot #3 +1.500 UP +2.000 UL +LT0 +0.50 0.00 0.50 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +2527 3443 M +(EP) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.50 0.00 0.50 C 1977 3443 M +298 0 V +725 1199 M +846 760 L +242 12 V +484 -94 V +969 534 V +4479 680 L +725 1199 Star +846 760 Star +1088 772 Star +1572 678 Star +2541 1212 Star +4479 680 Star +2126 3443 Star +% End plot #3 +% Begin plot #4 +1.500 UP +2.000 UL +LT0 +0.18 0.31 0.31 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +3161 3443 M +(LU) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.18 0.31 0.31 C 2611 3443 M +298 0 V +725 1739 M +846 676 L +242 1132 V +484 -709 V +969 211 V +4479 1082 L +725 1739 TriUF +846 676 TriUF +1088 1808 TriUF +1572 1099 TriUF +2541 1310 TriUF +4479 1082 TriUF +2760 3443 TriUF +% End plot #4 +% Begin plot #5 +1.500 UP +2.000 UL +LT0 +0.18 0.55 0.34 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +3795 3443 M +(BT) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.18 0.55 0.34 C 3245 3443 M +298 0 V +725 2140 M +876 2038 L +212 -362 V +606 6 V +847 1114 V +4964 896 L +725 2140 BoxF +876 2038 BoxF +1088 1676 BoxF +1694 1682 BoxF +2541 2796 BoxF +4964 896 BoxF +3394 3443 BoxF +% End plot #5 +% Begin plot #6 +1.500 UP +2.000 UL +LT0 +0.85 0.65 0.13 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +4429 3443 M +(SP) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.85 0.65 0.13 C 3879 3443 M +298 0 V +725 1658 M +876 1155 L +212 180 V +606 -64 V +2541 680 L +2423 3 V +725 1658 Circle +876 1155 Circle +1088 1335 Circle +1694 1271 Circle +2541 680 Circle +4964 683 Circle +4028 3443 Circle +% End plot #6 +% Begin plot #7 +1.500 UP +2.000 UL +LT0 +0.55 0.00 0.00 C LCb setrgbcolor +/Helvetica findfont 190 scalefont setfont +5063 3443 M +(FT) Rshow +/Helvetica findfont 140 scalefont setfont +LT0 +0.55 0.00 0.00 C 4513 3443 M +298 0 V +725 847 M +121 267 V +242 680 V +484 -617 V +969 339 V +4479 1166 L +725 847 CircleF +846 1114 CircleF +1088 1794 CircleF +1572 1177 CircleF +2541 1516 CircleF +4479 1166 CircleF +4662 3443 CircleF +% End plot #7 +1.000 UL +LTb +602 3611 N +602 588 L +4545 0 V +0 3023 V +-4545 0 V +Z stroke +1.000 UP +1.000 UL +LTb +stroke +grestore +end +showpage +%%Trailer +%%DocumentFonts: Helvetica diff --git a/fig/sen_comp.pdf b/fig/sen_comp.pdf new file mode 100644 index 0000000..f25c19b Binary files /dev/null and b/fig/sen_comp.pdf differ diff --git a/fig/start_freq.pdf b/fig/start_freq.pdf new file mode 100644 index 0000000..66c24cd Binary files /dev/null and b/fig/start_freq.pdf differ diff --git a/fig/three_scenarios.pdf b/fig/three_scenarios.pdf index 21b0524..9f36c43 100644 Binary files a/fig/three_scenarios.pdf and b/fig/three_scenarios.pdf differ diff --git a/my_reference.bib b/my_reference.bib index de5f548..032f405 100644 --- a/my_reference.bib +++ b/my_reference.bib @@ -798,33 +798,4 @@ ISSN={1045-9219},} url = {http://www.eia.gov/} } -@inproceedings{pdsec2015, - title = {Energy Consumption Reduction with DVFS for Message Passing Iterative Applications on Heterogeneous Architectures}, - author = {Charr, Jean-Claude and Couturier, Rapha\~{A}«l and Fanfakh, Ahmed and Giersch, Arnaud}, - year = {2015}, - address = {Hyderabad, India}, - booktitle = {PDSEC 2015, 16th IEEE Int. Workshop on Parallel and Distributed Scientific and Engineering Computing (in conjuction with IPDPS 2015)}, - month = {May}, - %pages = {***--***}, - publisher = {IEEE} -} - -@article{Energy_measurement, -year={2014}, -issn={0920-8542}, -journal={The Journal of Supercomputing}, -volume={70}, -number={3}, -doi={10.1007/s11227-014-1236-4}, -title={Energy measurement, modeling, and prediction for processors with frequency scaling}, -publisher={Springer US}, -keywords={Dynamic voltage–frequency scaling; DVFS; SPEC CPU2006 benchmarks; Energy measurement; Energy models}, -author={Rauber, Thomas and Rünger, Gudula and Schwind, Michael and Xu, Haibin and Melzner, Simon}, -pages={1451-1476} -} - - -@MISC{grid5000, - title = {grid5000}, - url = {http://www.grid5000.fr/} -} \ No newline at end of file + diff --git a/pdsec15_review.txt b/pdsec15_review.txt new file mode 100644 index 0000000..1b5dfdb --- /dev/null +++ b/pdsec15_review.txt @@ -0,0 +1,331 @@ +============================== Standard 1 ============================== + +> *** Key Contributions: Please describe the key contributions of the + paper or lack thereof. Your comments should be specific and + justify your overall recommendation. + +This paper presents a new online frequency selecting algorithm for +distributed iterative applications running on heterogeneous CPU nodes. +Contrary to previous work (for homogeneous CPU), this heterogeneous +context implies a vector of scaling factors and "slack times" before +synchronizing the processes at each iteration. The models and the +algorithm are clearly presented and detailed, and are validated on +several benchmarks thanks to a simulator. Comparison with another +scaling factor selection algorithm (which does not take into account +communication times and heterogeneity) shows the relevance of this new +algorithm which manages to significantly reduce the energy consumption +with acceptable performance overhead. + +Overall, this is a very solid work, and the paper is well-written and +very clear. + +The main flaw of this paper is that the evaluation is only done via a +simulator. As mentioned in future work, evaluations on real +heterogeneous CPU platforms (with real power measurements) will be +necessary (as future work) to validate definitely this algorithm and +the models. + +> *** Suggestions for Improvement: Additional comments and suggestions + for improvement in the technical content or the presentation. + Please be as detailed and constructive as you can be. + +The energy and performance models rely on compute-bound programs, +where the computation time is linearly proportional to the processor +frequency. Does this apply to all NAS benchmarks ? The authors should +specify which NAS benchmarks are memory-bound (if any), and how their +model apply to these memory-bound benchmarks. + +Moreover, in section III it seems that the authors assume that the +communication time (without slack time) is the same for all processors +provided they have the same communication volume. This could be +pointed out more clearly in the paper. Also, does this apply to all +NAS benchmarks? Does it also depends on the placement of the MPI +processes? I assume that for the same communication volume, the +communication time will differ whether the processes are on +neighbouring nodes or are on distant nodes (especially with 128 or 144 +nodes). +Could the authors discuss in the text? + +The authors consider that the communication time only apply to static +power, which means that no CPU cycle is used for the MPI +communications. Does this implies specific networks (like Infiniband) +with RDMA? +This could be clarified in the paper. + +Finally, the algorithm applies to synchronous iterative applications: +is this the case for all NAS benchmarks evaluated in this paper? This +could also be specified in the paper. + +Figures 2a and 2b : I do not understand why the energy curve in Fig.2b +does not have the same shape as the one in Fig.2a. +Could the authors specify this in the text? + +Minor comments : +- The authors could specify in the abstract that "heterogeneous + platforms" refer to heterogeneous CPUs (not to CPU-GPU nodes). +- The terms "in the same direction" (used twice in section IV) are + unclear and should be rewritten. +- Section V.A : replace "because selecting frequency scaling factors + higher than the higher bound" by "because selecting frequencies + higher than the higher bound"? + +> *** Significance: Assess the significance of the topic addressed in + the paper. + +Excellent (5) + +> *** Originality/Novelty (of contribution): How novel are the + concepts presented in the paper? + +Above average (4) + +> *** Technical Soundness: How strong are the techniques and + methodologies used in the paper? + +Excellent (5) + +> *** Overall Recommendation: Your final rating should be consistent + with your ratings on previous questions. + +Accept (5) + +============================== Standard 2 ============================== + +> *** Key Contributions: Please describe the key contributions of the + paper or lack thereof. Your comments should be specific and + justify your overall recommendation. + +The paper proposed a frequency selection algorithm for heterogeneous +platforms. The algorithm proposed the maximum distance between the +energy consumption and the performance to get the trade off scale +factor. on This is an interesting paper with good trial to cover many +factors. + +The paper ran NPB benchmarks to verify the algorithm but there is no +comparison between the results at the the trade-off scale factor and +those from all other possible scale factors without applying the +algorithm. Without this, it is not reliable to validate the algorithm. + +> *** Suggestions for Improvement: Additional comments and suggestions + for improvement in the technical content or the presentation. + Please be as detailed and constructive as you can be. + +There are too much tables i.e. II-VII in section VI. Better to +summarize them in a couple of figures. + +It is necessary to describe the overhead of the algorithm which is +missed in the paper. + +> *** Significance: Assess the significance of the topic addressed in + the paper. + +Average (3) + +> *** Originality/Novelty (of contribution): How novel are the + concepts presented in the paper? + +Average (3) + +> *** Technical Soundness: How strong are the techniques and + methodologies used in the paper? + +Acceptable (3) + +> *** Overall Recommendation: Your final rating should be consistent + with your ratings on previous questions. + +Weak Accept (4) + +============================== Standard 3 Thomas R. ============================== + +> *** Key Contributions: Please describe the key contributions of the + paper or lack thereof. Your comments should be specific and + justify your overall recommendation. + +The paper develops DVFS performance models and an online algorithm to +optimize time and energy for iterative message passing applications on +a heterogeneous CPU cluster. An objective function is developed to +express the time energy tradeoff. Results using a simulated framework +show worthwhile energy gains for acceptable loss of execution time. A +comparison with a more general pre-existing algorithm show modest +improvements in energy and and energy-time tradeoff. + +The paper is well-written and is technically sound. Its significance +is slightly diminished due to the fact that previous work has largely +dealt with this issue on scenarios that are of stronger interests +and/or are less specialized. + +> *** Suggestions for Improvement: Additional comments and suggestions + for improvement in the technical content or the presentation. + Please be as detailed and constructive as you can be. + +The abstract would be sharpened it it contained numbers relating to +the performance degradation and comparison. + +III.A. The modelling of the communication time being independent of +the frequency is questionable, even if it is backed up by a 10year old +reference. While slack time is not affected, my own research has shown +that communication bandwidth does clearly increase with frequency, +albeit in a sub-linear fashion. The use of taking the minimum for +communication time (3) needs better explanation, as it is +counter-intuitive. + +I would like to some explanation as to why it takes so many iterations +for the algorithm to select the best vector, and whether this can be +improved. While the NAS benchmarks have a standard number of +iterations, it would be helpful to the reader to indicate what these +are in VI. + +The results on a real heterogeneous platform in the future work will +be interesting. + +There are a number of small grammatical errors: + +p2. ``to satisfy some objectives while taking into account all the +constraints,'': a comma is needed before `while' to match the 2nd + +Fig2(b) normalize -> normalized + +p4 ``following the same direction'': use `follow' + +Alg1: F_diff_i: difference -> differences + +p6: on all left frequencies -> on all remaining frequencies + +while it lowers the frequency of all other nodes -> +while it lowers the frequencies of all other nodes + +``the proposed algorithm is not an exact method it does'': +put a : before it + +p8: on different number of nodes -> on different numbers of nodes +the GC benchmark significantly decrease -> +the CG benchmark significantly decreases + +> *** Significance: Assess the significance of the topic addressed in + the paper. + +Above average (4) + +> *** Originality/Novelty (of contribution): How novel are the + concepts presented in the paper? + +Above average (4) + +> *** Technical Soundness: How strong are the techniques and + methodologies used in the paper? + +Excellent (5) + +> *** Overall Recommendation: Your final rating should be consistent + with your ratings on previous questions. + +Strong Accept (6) + +============================== Standard 4 ============================== + +> *** Key Contributions: Please describe the key contributions of the + paper or lack thereof. Your comments should be specific and + justify your overall recommendation. + +In this paper, a new online frequency selecting algorithm for +heterogeneous platforms is presented. It selects the frequencies and +tries to give the best trade-off between energy saving and performance +degradation, for each node computing the message passing iterative +application. The algorithm has a small overhead and works without +training or profiling. It uses a new energy model for message passing +iterative applications running on a het- erogeneous platform. The +proposed algorithm is evaluated on the SimGrid simulator while running +the NAS parallel benchmarks. The experiments show that it reduces the +energy consumption by up to 35 % while limiting the performance +degradation as much as possible. Finally, the algorithm is compared to +an existing method, the comparison results showing that it outperforms +the latter. + +> *** Suggestions for Improvement: Additional comments and suggestions + for improvement in the technical content or the presentation. + Please be as detailed and constructive as you can be. + +I did not see every clearly that if the proposed online algorithm can +achieve the optimal selection. If only the heustrics, then how close +to the optimal? I would like to see more theoretical or experimental +results if possible since the authors claims the "the best trade-off +between energy saving and performance degradation". + +> *** Significance: Assess the significance of the topic addressed in + the paper. + +Excellent (5) + +> *** Originality/Novelty (of contribution): How novel are the + concepts presented in the paper? + +Excellent (5) + +> *** Technical Soundness: How strong are the techniques and + methodologies used in the paper? + +Strong (4) + +> *** Overall Recommendation: Your final rating should be consistent + with your ratings on previous questions. + +Strong Accept (6) + +============================== Standard 5 ============================== + +> *** Key Contributions: Please describe the key contributions of the + paper or lack thereof. Your comments should be specific and + justify your overall recommendation. + +The paper considers the DVFS technique and presents an energy model +for DVFS systems that also takes the communication time into +consideration. An new algorithm for selecting the scaling factors is +presented. The algorithm uses a vector of scaling factors, one for +each node, and determines the scaling factors such that best trade-off +between minimizing the energy consumption and maximizing the +performance for a synchronous iterative algorithm is reached. The +algorithm works during execution time and uses the first interation +step for collecting the information required for the scaling factor +selection. An experimental evaluation is given using the SimGrid +environment. + +The paper is well written and structured and should be accepted. It +is solid work and provides new contributions by extending earlier +energy models with communication time concerns and proposes a new +algorithm for DVFS control. + +> *** Suggestions for Improvement: Additional comments and suggestions + for improvement in the technical content or the presentation. + Please be as detailed and constructive as you can be. + +Algorithm 1 in Section V could be explained in more detail. As far as +I can see, it tests all possible frequencies or scaling factors for +the different nodes and selects the best as indicated by the model. I +was wondering whether all combinations of scaling factors are tested +or whether this is not necessary because of the behavior of the +communication. +The accuracy of the frequency selection depends on the accuracy of the +model used for the computation of the scaling factors. It would be +interesting to see how accurate the model is for real systems. +However, I see that this might be difficult to capture in practice. + +> *** Significance: Assess the significance of the topic addressed in + the paper. + +Excellent (5) + +> *** Originality/Novelty (of contribution): How novel are the + concepts presented in the paper? + +Above average (4) + +> *** Technical Soundness: How strong are the techniques and + methodologies used in the paper? + +Excellent (5) + +> *** Overall Recommendation: Your final rating should be consistent + with your ratings on previous questions. + +Accept (5)