From: Kahina Date: Mon, 21 Dec 2015 06:21:14 +0000 (+0100) Subject: MAJ X-Git-Url: https://bilbo.iut-bm.univ-fcomte.fr/and/gitweb/kahina_paper2.git/commitdiff_plain/d1d299f31705641d990ab7b8d31f40ae5299144d?ds=sidebyside MAJ --- diff --git a/paper.tex b/paper.tex index 88d8823..8d0dc13 100644 --- a/paper.tex +++ b/paper.tex @@ -492,7 +492,14 @@ We introduced three paradigms of parallel programming. Our objective consist to \section{The EA algorithm on single GPU} \subsection{the EA method} -the Ehrlich-Aberth method is an iterative method , contain 4 steps, start from the initial approximations of all the +%\begin{figure}[htbp] +%\centering + % \includegraphics[angle=-90,width=0.5\textwidth]{EA-Algorithm} +%\caption{The Ehrlich-Aberth algorithm on single GPU} +%\label{fig:03} +%\end{figure} + +the Ehrlich-Aberth method is an iterative method, contain 4 steps, start from the initial approximations of all the roots of the polynomial,the second step initialize the solution vector $Z$ using the Guggenheimer method to assure the distinction of the initial vector roots, than in step 3 we apply the the iterative function based on the Newton's method and Weiestrass operator[...,...], wich will make it possible to converge to the roots solution, provided that all the root are different. At the end of each application of the iterative function, a stop condition is verified consists in stopping the iterative process when the whole of the modules of the roots are lower than a fixed value $ε$ \subsection{EA parallel implementation on CUDA} @@ -567,6 +574,12 @@ Algorithm~\ref{alg2-cuda} shows a sketch of the Ehrlich-Aberth method using CUDA \subsection{MGPU (OpenMP-CUDA)approach} Before starting computations, our parallel implementation shared input data of the root finding polynomial between OpenMP threads. From Algorithm 1, the input data are the solution vector $Z$, the polynomial to solve $P$. Let number of OpenMP threads is equal to the number of GPUs, each threads OpenMP ( T-omp) checks one GPU, and control a part of the shared memory, that is a part of the vector Z like: $(n/Nbr_gpu)$ roots, n: the polynomial's degrees, $Nbr_gpu$ the number of GPUs. Then every GPU will have a grid of computation organized with its performances and the size of data of which it checks. In principle a grid is set by two parameter DimGrid, the number of block per grid, DimBloc: the number of threads per block. The following schema shows the architecture of (CUDA,OpenMP). +%\begin{figure}[htbp] +%\centering + % \includegraphics[angle=-90,width=0.5\textwidth]{OpenMP-CUDA} +%\caption{The OpenMP-CUDA architecture} +%\label{fig:03} +%\end{figure} Each thread OpenMP compute the kernels on GPUs,than after each iteration they copy out the data from GPU memory to CPU shared memory. The kernels are re-runs is up to the roots converge sufficiently. Here are below the corresponding algorithm: @@ -574,43 +587,83 @@ Each thread OpenMP compute the kernels on GPUs,than after each iteration they co \begin{algorithm}[htpb] \label{alg2-cuda} %\LinesNumbered -\caption{CUDA OpenMP Algorithm to find roots with the Ehrlich-Aberth method} +\caption{CUDA-OpenMP Algorithm to find roots with the Ehrlich-Aberth method} \KwIn{$Z^{0}$ (Initial root's vector), $\varepsilon$ (Error tolerance - threshold), P (Polynomial to solve), Pu (Derivative of P), $n$ (Polynomial degrees), $\Delta z_{max}$ (Maximum value of stop condition)} + threshold), P (Polynomial to solve), Pu (Derivative of P), $n$ (Polynomial degrees), $\Delta z$ ( Vector of errors of stop condition), $num_gpus$ (number of OpenMP threads/ number of GPUs), Size (number of roots)} \KwOut {$Z$ (Solution root's vector), $ZPrec$ (Previous solution root's vector)} \BlankLine -// selection du GPU\; -\item cudaSetDevice(i)\; -// allocations memoire\; -\verb= #pragma omp single= -\item hostAlloc(P,Pu,Z)\; -\verb= #pragma omp parallel shared(Z,∆zmax,P)= -\item deviceAlloc(dP,dPu,dZ)\; -\verb= #pragma omp barrier= -// transfers CPU-GPU and compute GPU\; -\item copyH2D(P,dP)\; -\item copyH2D(Pu,dPu)\; -\item copyH2D(Zi,dZi)\; -\While {$\Delta z_{max} > \epsilon$}{ -\item Let $\Delta z_{max}=0$\; + +\item Initialization of the of P\; +\item Initialization of the of Pu\; +\item Initialization of the solution vector $Z^{0}$\; +\verb=omp_set_num_threads(num_gpus);= +\verb=cudaGetDevice(gpu_id);= +\verb=#pragma omp parallel shared(Z,$\Delta$ z,P);= +\item Allocate and copy initial data from CPU memory to the GPU global memories\; +\item index= $Size/num\_gpus$\; +\item k=0\; +\While {$error > \epsilon$}{ +\item Let $\Delta z=0$\; \item $ kernel\_save(ZPrec,Z)$\; \item k=k+1\; -//each GPU i compute the new root for his part dZi -\item $ kernel\_update(dZi,P,Pu)$\; -\item $kernel\_testConverge(\Delta z_{max},dZi,ZPrec)$\; +\item $ kernel\_update(Z,P,Pu,index)$\; +\item $kernel\_testConverge(\Delta z[gpu\_id],Z,ZPrec)$\; +%\verb=#pragma omp barrier;= +\item error= Max($\Delta z$)\; } -\item copyD2H(dZ,Zi)\; - // fin omp parallel\; +\item Copy results from GPU memories to CPU memory\; \end{algorithm} \end{enumerate} ~\\ -\subsection{MGPU (MPI-CUDA)approach} + +\subsection{Multi-GPU (MPI-CUDA)approach} +%\begin{figure}[htbp] +%\centering + % \includegraphics[angle=-90,width=0.2\textwidth]{MPI-CUDA} +%\caption{The MPI-CUDA architecture } +%\label{fig:03} +%\end{figure} + + +\begin{enumerate} +\begin{algorithm}[htpb] +\label{alg2-cuda} +%\LinesNumbered +\caption{CUDA-MPI Algorithm to find roots with the Ehrlich-Aberth method} + +\KwIn{$Z^{0}$ (Initial root's vector), $\varepsilon$ (Error tolerance + threshold), P (Polynomial to solve), Pu (Derivative of P), $n$ (Polynomial degrees), $\Delta z$ ( error of stop condition), $num_gpus$ (number of MPI processes/ number of GPUs), Size (number of roots)} + +\KwOut {$Z$ (Solution root's vector), $ZPrec$ (Previous solution root's vector)} + +\BlankLine +\item Initialization of the P\; +\item Initialization of the Pu\; +\item Initialization of the solution vector $Z^{0}$\; +\item Allocate and copy initial data from CPU memories to the GPU global memories\; +\item $index= Size/num_gpus$\; +\item k=0\; +\While {$error > \epsilon$}{ +\item Let $\Delta z=0$\; +\item $ kernel\_save(ZPrec,Z)$\; +\item k=k+1\; +\item $ kernel\_update(Z,P,Pu,index)$\; +\item $kernel\_testConverge(\Delta z,Z,ZPrec)$\; +\item Copy results from GPU memories to CPU memories\; +\item Send $Z[id]$ to all neighboring processes\; +\item Receive $Z[j]$ from neighboring process j\; +\item ComputeMaxError($\Delta z$,error)\; + +} +\end{algorithm} +\end{enumerate} +~\\ \section{experiments}