X-Git-Url: https://bilbo.iut-bm.univ-fcomte.fr/and/gitweb/mpi-energy2.git/blobdiff_plain/c9875e5d70672da61c5cdc75fac90dad56cb1b04..dc41bd6acde81b12c7e48075d1a74dee1f30db73:/Heter_paper.tex

diff --git a/Heter_paper.tex b/Heter_paper.tex
index 2fc8549..a6d44b1 100644
--- a/Heter_paper.tex
+++ b/Heter_paper.tex
@@ -76,7 +76,19 @@
 \maketitle
 
 \begin{abstract}
-  
+Green computing emphasizes the importance of energy conservation, minimizing the negative impact 
+on the environment while achieving high performance and minimizing operating costs. So, energy reduction 
+process in a high performance clusters it can be archived using dynamic voltage and frequency 
+scaling (DVFS) technique, through reducing the frequency of a CPU. Using DVFS to lower levels 
+result in a high increase in performance degradation ratio. Therefore selecting the best frequencies 
+must give the best possible tradeoff between the energy and the performance of parallel program.
+
+In this paper we present a new online heterogeneous scaling algorithm that selects the best vector 
+of frequency scaling factors. These factors give the best tradeoff between the energy saving and the
+performance degradation. The algorithm has small overhead and works without training and profiling.
+We developed a new energy model for distributed iterative application running on heterogeneous cluster. 
+The proposed algorithm experimented  on Simgrid simulator that applying the NAS parallel benchmarks.
+It reduces the energy consumption up to 35\% while limits the performance degradation as much as possible.
 \end{abstract}
 
 \section{Introduction}
@@ -153,9 +165,9 @@ on  CPUs. As an example of this works, Luley et al.
 cluster composed of Intel Xeon CPUs and NVIDIA GPUs. Their main goal is to determined the 
 energy efficiency as a function of performance per watt, the best tradeoff is done when the 
 performance per watt function is maximized. In the work of Kia Ma et al.
-~\cite{KaiMa_Holistic.Approach.to.Energy.Efficiency.in.GPU-CPU}, They developed a scheduling 
+~\cite{KaiMa_Holistic.Approach.to.Energy.Efficiency.in.GPU-CPU}, they developed a scheduling 
 algorithm to distributed different workloads proportional to the computing power of the node 
-to be executed on a CPU or a GPU, emphasize all tasks must be finished in the same time. 
+to be executed on CPU or GPU, emphasize all tasks must be finished in the same time. 
 Recently, Rong et al.~\cite{Rong_Effects.of.DVFS.on.K20.GPU}, Their study explain that 
 a heterogeneous clusters enabled DVFS using GPUs and CPUs gave better energy and performance 
 efficiency than other clusters composed of only CPUs. 
@@ -910,7 +922,7 @@ down the frequencies of some nodes have 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 80\% composed of  dynamic power and 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  
@@ -1040,15 +1052,29 @@ for a heterogeneous cluster composed of four different types of nodes having the
 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 $O(F\cdot (N \cdot4) )$, 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 section (\ref{sec.res}).
+vector of frequency scaling factors that gives the results of the sections (\ref{sec.res}) and (\ref{sec.compare}) .
 
 \section{Conclusion}
 \label{sec.concl}
-
+In this paper, we have presented a new online heterogeneous scaling algorithm
+that selects the best possible vector of frequency scaling factors. This vector 
+gives the maximum distance (optimal tradeoff) between the normalized energy and 
+the performance curves. In addition, we developed a new energy model for measuring  
+and predicting the energy of distributed iterative applications running over heterogeneous 
+cluster. The proposed method evaluated on Simgrid/SMPI  simulator to built a heterogeneous 
+platform to executes NAS parallel benchmarks. The results of the experiments showed the ability of
+the proposed algorithm to changes its behaviour to selects different scaling factors  when 
+the number of computing nodes and both of the static and the dynamic powers are changed. 
+
+In the future, we plan to improve this method to apply on asynchronous  iterative applications 
+where each task does not wait the others tasks to finish there works. This leads us to develop a new 
+energy model to an asynchronous iterative applications, where the number of iterations is not 
+known in advance and depends on the global convergence of the iterative system.
 
 \section*{Acknowledgment}
 
 
+
 % trigger a \newpage just before the given reference
 % number - used to balance the columns on the last page
 % adjust value as needed - may need to be readjusted if