small modifs

[mpi-energy2.git] / Heter_paper.tex

diff --git a/Heter_paper.tex b/Heter_paper.tex
index d31891279930b9743c2e58c7b367fefb4c3a4c37..c29b2a40b302fc1d68f2a707dd36ab2da9534c32 100644 (file)
--- a/Heter_paper.tex
+++ b/Heter_paper.tex
@@ -89,7 +89,8 @@ for each node computing the message passing iterative application. The algorithm
 works without training or profiling. It uses a new energy model for message passing iterative applications 
 running on a heterogeneous platform. The proposed algorithm is evaluated  on the Simgrid simulator while 
 running the NAS parallel benchmarks. The experiments demonstrated that it reduces the energy consumption 
 works without training or profiling. It uses a new energy model for message passing iterative applications 
 running on a heterogeneous platform. The proposed algorithm is evaluated  on the Simgrid simulator while 
 running the NAS parallel benchmarks. The experiments demonstrated that it reduces the energy consumption 

-up to 35\% while limiting the performance degradation as much as possible.
+up to 35\% while limiting the performance degradation as much as possible. Finally, the algorithm is compared to an existing method and the comparison results show that it outperforms the latter.
+

 \end{abstract}
 
 \section{Introduction}
 \end{abstract}
 
 \section{Introduction}


@@ -109,7 +110,7 @@ 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 consumed 57.15 kilowatts.
 
 became the top of the Green500 list in November 2014 \cite{Green500_List}. 
 This heterogeneous platform executes more than 5  GFLOPS per watt while consumed 57.15 kilowatts.
 

-Besides hardware improvements, there are many software techniques to lower the energy consumption of these platforms, 
+Besides platform improvements, there are many software and hardware techniques to lower the energy consumption of these platforms, 

 such as scheduling, DVFS, ... 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 might increase  the execution time of the application running over that processor.
 such as scheduling, DVFS, ... 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 might increase  the execution time of the application running over that processor.


@@ -127,8 +128,9 @@ the energy-performance objective function that maximizes the reduction of energy
 consumption while minimizing the degradation of the program's performance.
 Section~\ref{sec.optim} details the proposed frequency selecting algorithm then the precision of the proposed algorithm is verified. 
 Section~\ref{sec.expe} presents the results of applying the algorithm on  the NAS parallel benchmarks and executing them 
 consumption while minimizing the degradation of the program's performance.
 Section~\ref{sec.optim} details the proposed frequency selecting algorithm then the precision of the proposed algorithm is verified. 
 Section~\ref{sec.expe} presents the results of applying the algorithm on  the NAS parallel benchmarks and executing them 

-on a heterogeneous platform. It also shows the results of running three 
-different power scenarios and comparing them. 
+on a heterogeneous platform. It shows the results of running three 
+different power scenarios and comparing them. Moreover, it also shows the comparison results
+between the proposed method and an existing method.

 Finally, in Section~\ref{sec.concl} the paper is ended with a summary and some future works.
 
 \section{Related works}
 Finally, in Section~\ref{sec.concl} the paper is ended with a summary and some future works.
 
 \section{Related works}


@@ -136,7 +138,7 @@ Finally, in Section~\ref{sec.concl} the paper is ended with a summary and some f
 DVFS is a technique enabled 
 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 
 DVFS is a technique enabled 
 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 the GPUs to achieve the same goal. Reducing the frequency of a processor lowers its number of FLOPS and might 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 and 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,Dhiman_Online.Learning.Power.Management}. Others used offline methods that might 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, ... 
+also  allowed in the GPUs to achieve the same goal. Reducing the frequency of a processor lowers its number of FLOPS and might 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 and 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 might 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, ... 

 
 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: 
 
 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: 


@@ -207,7 +209,7 @@ task which have the highest computation time and no slack time.
   
  \begin{figure}[t]
   \centering
   
  \begin{figure}[t]
   \centering

-   \includegraphics[scale=0.5]{fig/commtasks}
+   \includegraphics[scale=0.6]{fig/commtasks}

   \caption{Parallel tasks on a heterogeneous platform}
   \label{fig:heter}
 \end{figure}
   \caption{Parallel tasks on a heterogeneous platform}
   \label{fig:heter}
 \end{figure}


@@ -252,7 +254,7 @@ vector of scaling factors can be predicted using (\ref{eq:perf}).
 \end{equation}
 Where:\\
 \begin{equation}
 \end{equation}
 Where:\\
 \begin{equation}

-\label{eq:perf}
+\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 
  MinTcm = \min_{i=1,2,\dots,N} (Tcm_i)
 \end{equation}
 where $TcpOld_i$ is the computation time  of processor $i$ during the first 


@@ -266,7 +268,7 @@ by the number of iterations of that application.
 
 This prediction model is developed from the model for predicting the execution time of 
 message passing distributed applications for homogeneous architectures~\cite{Our_first_paper}. 
 
 This prediction model is developed from the model for predicting the execution time of 
 message passing distributed applications for homogeneous architectures~\cite{Our_first_paper}. 

-The execution time prediction model is uSpiliopoulossed in the method for optimizing both 
+The execution time prediction model is used in the method for optimizing both 

 energy consumption and performance of iterative methods, which is presented in the 
 following sections.
 
 energy consumption and performance of iterative methods, which is presented in the 
 following sections.
 


@@ -356,7 +358,7 @@ The communication time of a processor $i$ is noted as $Tcm_{i}$ and could contai
 if it is 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 
 if it is 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 each processor multiplied by its static power. 
+computed as the sum of the execution time of one iteration multiplied by static power of each processor. 

 The overall energy consumption of a message passing  distributed application executed over a 
 heterogeneous platform during one iteration is the summation of all dynamic and static energies 
 for each  processor.  It is computed as follows:
 The overall energy consumption of a message passing  distributed application executed over a 
 heterogeneous platform during one iteration is the summation of all dynamic and static energies 
 for each  processor.  It is computed as follows:


@@ -433,7 +435,7 @@ 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 
 reduction with minimum execution time reduction.  
   
 This problem can be solved by making the optimization process for energy and 

-execution time follow the same direction.  Therefore, the equation of the 
+execution time following the same direction.  Therefore, the equation of the 

 normalized execution time is inverted which gives the normalized performance equation, as follows:
 \begin{multline}
   \label{eq:pnorm_inv}
 normalized execution time is inverted which gives the normalized performance equation, as follows:
 \begin{multline}
   \label{eq:pnorm_inv}


@@ -478,14 +480,14 @@ the energy curve has a convex form as shown in~\cite{Zhuo_Energy.efficient.Dynam
 \label{sec.optim}
 
 \subsection{The algorithm details}
 \label{sec.optim}
 
 \subsection{The algorithm details}

-In this section algorithm \ref{HSA} is presented. It selects the frequency scaling factors 
+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 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 executed  after the first 
 iteration and returns a vector of optimal frequency scaling factors   that satisfies the objective 
 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 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 executed  after the first 
 iteration and returns a vector of optimal frequency scaling factors   that satisfies the objective 

-function (\ref{eq:max}). The program apply DVFS operations to change the frequencies of the CPUs 
+function (\ref{eq:max}). The program applies DVFS operations to change the frequencies of the CPUs 

 according to the computed scaling factors.  This algorithm is called just once during the execution 
 of the program. Algorithm~(\ref{dvfs}) shows where and when the proposed scaling algorithm is called 
 in the iterative MPI program.
 according to the computed scaling factors.  This algorithm is called just once during the execution 
 of the program. Algorithm~(\ref{dvfs}) shows where and when the proposed scaling algorithm is called 
 in the iterative MPI program.


@@ -524,7 +526,7 @@ scaling factors starts the search method from these initial frequencies and take
 toward lower frequencies. The algorithm iterates on all left 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 
 toward lower frequencies. The algorithm iterates on all left 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 (\ref{eq:perf}) and keeps its frequency unchanged, while it lowers the frequency of  
+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 factors. 
 The optimal set of frequency scaling factors is the set that gives the highest distance according to  the objective 
 all other nodes by one gear.
 The new overall energy consumption and execution time are computed according to the new scaling factors. 
 The optimal set of frequency scaling factors is the set that gives the highest distance according to  the objective 


@@ -620,7 +622,7 @@ which results in bigger energy savings.
 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 
 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 the execution time and the dynamic energy is related to the computation time. So, all of 
+linearly related to the execution time and the dynamic energy is related to the computation time. So, all of 

 the works presented in this paper is 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 
 the works presented in this paper is 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 


@@ -1018,10 +1020,10 @@ results in less energy saving but less performance degradation.
 \begin{figure}
   \centering
   \subfloat[Comparison  of the results on 8 nodes]{%
 \begin{figure}
   \centering
   \subfloat[Comparison  of the results on 8 nodes]{%

-    \includegraphics[width=.30\textwidth]{fig/sen_comp}\label{fig:sen_comp}}%
+    \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]{%
 
   \subfloat[Comparison the selected frequency scaling factors of MG benchmark class C running on 8 nodes]{%

-    \includegraphics[width=.34\textwidth]{fig/three_scenarios}\label{fig:scales_comp}}
+    \includegraphics[width=.33\textwidth]{fig/three_scenarios}\label{fig:scales_comp}}

   \label{fig:comp}
   \caption{The comparison of the three power scenarios}
 \end{figure}  
   \label{fig:comp}
   \caption{The comparison of the three power scenarios}
 \end{figure}  


@@ -1031,27 +1033,20 @@ results in less energy saving but less performance degradation.
 
 \subsection{The comparison of the proposed scaling algorithm }
 \label{sec.compare_EDP}
 
 \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, $EDP=Enegry*Delay$ 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 consumptions 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 gives better energy savings than Spiliopoulos et al. algorithm, 
+on average it results in 29.76\% energy saving while their algorithm returns just 25.75\%. The average of performance degradation percentage is approximately the same for both algorithms, about 4\%. 
+
+
+For all benchmarks, our algorithm outperforms 
+Spiliopoulos et al. algorithm in term of energy and performance tradeoff, 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. 
+

 
 

-In this section, we compare our scaling  factors selection algorithm
-with Spiliopoulos et al. algorithm \cite{Spiliopoulos_Green.governors.Adaptive.DVFS}. 
-They developed an online frequency selecting algorithm running over multicore architecture. 
-The algorithm predicted both the energy and performance during the runtime of the program, then 
-selecting the frequencies that minimized the energy and delay products (EDP), $EDP=Enegry * Delay$. 
-To be able to compare with this algorithm, we used our energy and execution time models in prediction process,
-equations (\ref{eq:energy}) and  (\ref{eq:fnew}). Also their algorithm is adapted to taking into account 
-the heterogeneous platform to starts selecting the 
-initial frequencies using the equation (\ref{eq:Fint}). The algorithm built to test all possible frequencies as 
-a brute-force search algorithm. 
-
-The comparison results of running NAS benchmarks class C on 8 or 9 nodes are 
-presented in table \ref{table:compare_EDP}. The results show that our algorithm has a biggest energy saving percentage, 
-on average it has 29.76\% and thier algorithm has 25.75\%,
-while the average of performance degradation percentage is approximately the same, the average for our algorithm is 
-equal to 3.89\% and for their algorithm is equal to 4.03\%. In general, our algorithm outperforms 
-Spiliopoulos et al. algorithm in term of energy and performance tradeoff see figure (\ref{fig:compare_EDP}). 
-This because our algorithm maximized the difference (the distance) between the energy saving and the performance degradation 
-comparing to their EDP optimization function. It is also keeps the frequency of the slowest node without change 
-that gave some enhancements to the energy and performance tradeoff.

 
 
 \begin{table}[h]
 
 
 \begin{table}[h]


@@ -1075,9 +1070,11 @@ that gave some enhancements to the energy and performance tradeoff.
 
 
 
 
 
 

+
+

 \begin{figure}[t]
   \centering
 \begin{figure}[t]
   \centering

-   \includegraphics[scale=0.6]{fig/compare_EDP.pdf}
+   \includegraphics[scale=0.5]{fig/compare_EDP.pdf}

   \caption{Tradeoff comparison for NAS benchmarks class C}
   \label{fig:compare_EDP}
 \end{figure}
   \caption{Tradeoff comparison for NAS benchmarks class C}
   \label{fig:compare_EDP}
 \end{figure}


@@ -1088,19 +1085,20 @@ that gave some enhancements to the energy and performance tradeoff.
 In this paper, a new online frequency selecting algorithm has been presented. It selects the best possible vector of frequency scaling factors that gives the maximum distance (optimal tradeoff) between the predicted energy and 
 the predicted performance curves for a heterogeneous platform. This algorithm uses a new energy model for measuring  
 and predicting the energy of distributed iterative applications running over heterogeneous 
 In this paper, a new online frequency selecting algorithm has been presented. It selects the best possible vector of frequency scaling factors that gives the maximum distance (optimal tradeoff) between the predicted energy and 
 the predicted performance curves for a heterogeneous platform. This algorithm uses a new energy model for measuring  
 and predicting the energy of distributed iterative applications running over heterogeneous 

-platform. To evaluate the proposed method, it  was  applied on the NAS parallel benchmarks and executed over a heterogeneous platform simulated by  Simgrid. The results of the experiments showed that the algorithm reduces up to 35\% the energy consumption of a message passing iterative method while limiting the degradation of the performance. The algorithm also  selects different scaling factors   according to the percentage of the computing and communication times, and according to the values of  the static and  dynamic powers of the CPUs. 
+platform. To evaluate the proposed method, it  was  applied on the NAS parallel benchmarks and executed over a heterogeneous platform simulated by  Simgrid. The results of the experiments showed that the algorithm reduces up to 35\% the energy consumption of a message passing iterative method while limiting the degradation of the performance. The algorithm also  selects different scaling factors   according to the percentage of the computing and communication times, and according to the values of  the static and  dynamic powers of the CPUs. Finally, the algorithm was compared to Spiliopoulos et al. algorithm and the results showed that it 
+ outperforms their algorithm in term of energy-time tradeoff.

 
 In the near future, this method will be applied to real heterogeneous platforms to evaluate its performance in a real study case. It would also be interesting to evaluate its scalability over large scale heterogeneous platform and measure the energy consumption reduction it can produce. Afterward, we would like  to develop a similar method that is adapted to asynchronous  iterative applications 
 
 In the near future, this method will be applied to real heterogeneous platforms to evaluate its performance in a real study case. It would also be interesting to evaluate its scalability over large scale heterogeneous platform and measure the energy consumption reduction it can produce. Afterward, we would like  to develop a similar method that is adapted to asynchronous  iterative applications 

-where each task does not wait for others tasks to finish there works. The development of such method might require a new 
+where each task does not wait for others tasks to finish their works. The development of such 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}
 
 This work has been partially supported by the Labex
 energy model because the number of iterations is not 
 known in advance and depends on the global convergence of the iterative system.
 
 \section*{Acknowledgment}
 
 This work has been partially supported by the Labex

-ACTION project (contract “ANR-11-LABX-01-01”). As a PhD student,
+ACTION project (contract “ANR-11-LABX-01-01”). As a PhD student, 

 Mr. Ahmed Fanfakh, would like to thank the University of
 Mr. Ahmed Fanfakh, would like to thank the University of

-Babylon (Iraq) for supporting his work.
+Babylon (Iraq) for supporting his work. 

 
 
 % trigger a \newpage just before the given reference
 
 
 % trigger a \newpage just before the given reference


@@ -1112,7 +1110,7 @@ Babylon (Iraq) for supporting his work.
 \bibliographystyle{IEEEtran}
 \bibliography{IEEEabrv,my_reference}
 \end{document}
 \bibliographystyle{IEEEtran}
 \bibliography{IEEEabrv,my_reference}
 \end{document}

-
+   

 %%% Local Variables:
 %%% mode: latex
 %%% TeX-master: t
 %%% Local Variables:
 %%% mode: latex
 %%% TeX-master: t