Add review from PDSEC 2015.

[mpi-energy2.git] / Heter_paper.tex

diff --git a/Heter_paper.tex b/Heter_paper.tex
index 3ef5f181d4ef2296e1cf15e723a1cea3233ed27c..a9a1e6da497ddbd13099d0c00c389ec2b6a52213 100644 (file)
--- a/Heter_paper.tex
+++ b/Heter_paper.tex
@@ -59,7 +59,9 @@
 
 \begin{document}
 
 
 \begin{document}
 

-\title{Energy Consumption Reduction for Message Passing Iterative  Applications in Heterogeneous Architecture Using DVFS}
+\title{Energy Consumption Reduction with DVFS for \\
+  Message Passing Iterative Applications on \\
+  Heterogeneous Architectures}

 
 \author{%
   \IEEEauthorblockN{%
 
 \author{%
   \IEEEauthorblockN{%


@@ -69,7 +71,7 @@
     Arnaud Giersch
   }
   \IEEEauthorblockA{%
     Arnaud Giersch
   }
   \IEEEauthorblockA{%

-    FEMTO-ST Institute, University of Franche-Comte\\
+    FEMTO-ST Institute, University of Franche-Comté\\

     IUT de Belfort-Montbéliard,
     19 avenue du Maréchal Juin, BP 527, 90016 Belfort cedex, France\\
     % Telephone: \mbox{+33 3 84 58 77 86}, % Raphaël
     IUT de Belfort-Montbéliard,
     19 avenue du Maréchal Juin, BP 527, 90016 Belfort cedex, France\\
     % Telephone: \mbox{+33 3 84 58 77 86}, % Raphaël


@@ -85,7 +87,7 @@
   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
   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 might increase
+  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.
   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.


@@ -110,7 +112,7 @@
 
 The need for more computing power is continually increasing. To partially
 satisfy this need, most supercomputers constructors just put more computing
 
 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 might achieve higher floating
+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
 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


@@ -131,7 +133,7 @@ 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
 \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 might increase the execution
+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
 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


@@ -166,7 +168,7 @@ in Section~\ref{sec.concl} the paper ends with a summary and some future works.
 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
 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 might
+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
 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


@@ -174,7 +176,7 @@ 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}.
 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
+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
 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


@@ -268,7 +270,7 @@ have the same network bandwidth and latency.
 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
 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 nodes, slack times might 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
 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


@@ -278,7 +280,7 @@ Dynamic Voltage and Frequency Scaling (DVFS) is a process, implemented in
 modern processors, that reduces the energy consumption of a CPU by scaling
 down its voltage and frequency.  Since DVFS lowers the frequency of a CPU
 and consequently its computing power, the execution time of a program running
 modern processors, that reduces the energy consumption of a CPU by scaling
 down its voltage and frequency.  Since DVFS lowers the frequency of a CPU
 and consequently its computing power, the execution time of a program running

-over that scaled down processor might increase, especially if the program is
+over that scaled down processor may increase, especially if the program is

 compute bound.  The frequency reduction process can be  expressed by the scaling
 factor S which is the ratio between  the maximum and the new frequency of a CPU
 as in (\ref{eq:s}).
 compute bound.  The frequency reduction process can be  expressed by the scaling
 factor S which is the ratio between  the maximum and the new frequency of a CPU
 as in (\ref{eq:s}).


@@ -413,11 +415,11 @@ processor after scaling its frequency is computed as follows:
   \Es = \Ps \cdot (\Tcp \cdot S  + \Tcm)
 \end{equation}
 
   \Es = \Ps \cdot (\Tcp \cdot S  + \Tcm)
 \end{equation}
 

-In the considered heterogeneous platform, each processor $i$ might have
+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
 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

-$\Tcp[i]$ might be different and different frequency scaling factors might be
+$\Tcp[i]$ 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 $i$ is noted as
 $\Tcm[i]$ and could contain slack times when communicating with slower nodes,
 computed in order to decrease the overall energy consumption of the application
 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,


@@ -729,9 +731,9 @@ 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
 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

-number of iterations 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.
+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}
 
 \begin{table}[!t]
   \caption{Heterogeneous nodes characteristics}


@@ -778,7 +780,7 @@ 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
 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 1 Gbit/s bandwidth.
+nodes were connected via an Ethernet network with \np[Gbit/s]{1} bandwidth.

 
 
 \subsection{The experimental results of the scaling algorithm}
 
 
 \subsection{The experimental results of the scaling algorithm}


@@ -923,7 +925,7 @@ to be executed on 1, 4, 9, 16, 36, 64, or 144 nodes.
     \hline
     MG      &  3.27     &   1534.70     & 29.27    & 14.35         & 14.92    \\
     \hline
     \hline
     MG      &  3.27     &   1534.70     & 29.27    & 14.35         & 14.92    \\
     \hline

-    EP      &  5.05     &   5471.1084   & 27.12    &  3.11         & 24.01    \\
+    EP      &  5.05     &   5471.11     & 27.12    &  3.11         & 24.01    \\

     \hline
     LU      & 73.92     & 101339.16     & 21.96    &  3.67         & 18.29    \\
     \hline
     \hline
     LU      & 73.92     & 101339.16     & 21.96    &  3.67         & 18.29    \\
     \hline


@@ -1002,11 +1004,11 @@ 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,
 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 64GB) 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
+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.
 
 significantly increased when the number of computing nodes is increased because
 the computation times are small when compared to the communication times.
 


@@ -1117,7 +1119,7 @@ degradation.
     \hline
     CG      &  2812.38        & 36.36       &  6.80         & 29.56    \\
     \hline
     \hline
     CG      &  2812.38        & 36.36       &  6.80         & 29.56    \\
     \hline

-    MG      &   825.427       & 38.35       &  6.41         & 31.94    \\
+    MG      &   825.43        & 38.35       &  6.41         & 31.94    \\

     \hline
     EP      &  5281.62        & 35.02       &  2.68         & 32.34    \\
     \hline
     \hline
     EP      &  5281.62        & 35.02       &  2.68         & 32.34    \\
     \hline


@@ -1148,7 +1150,7 @@ degradation.
     \hline
     MG      & 29.49 & 33.78   & 3.74 & 6.41    & 25.75 & 27.37   \\
     \hline
     \hline
     MG      & 29.49 & 33.78   & 3.74 & 6.41    & 25.75 & 27.37   \\
     \hline

-    LU      & 19.55 & 28.33   & 0.0  & 0.01    & 19.55 & 28.22   \\
+    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
     \hline
     EP      & 28.40 & 27.04   & 4.29 & 0.49    & 24.11 & 26.55   \\
     \hline


@@ -1247,7 +1249,7 @@ the iterative system.
 \section*{Acknowledgment}
 
 This work has been partially supported by the Labex
 \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
 Babylon (Iraq) for supporting his work.
 
 Mr. Ahmed Fanfakh, would like to thank the University of
 Babylon (Iraq) for supporting his work.