branching instructions are used (if, while, ...), the better performance is
obtained on GPU. So with algorithm \ref{algo:seqCIprng} presented in the
previous section, it is possible to build a similar program which computes PRNG
-on GPU. The principe consists in assigning the computation of a PRNG as in
-sequential to each thread of the GPU. Of course, it is essential that the three
-xor-like PRNGs used for our computation have different parameters. So we chose
-them randomly with another PRNG. As the initialisation is performed by the CPU,
-we have chosen to use the ISAAC PRNG [ref] to initalize all the parameters for
-the GPU version of our PRNG. The implementation of the three xor-like PRNGs is
+on GPU.
+
+
+\subsection{Naive version}
+
+From the CPU version, it is possible to obtain a quite similar version for GPU.
+The principe consists in assigning the computation of a PRNG as in sequential to
+each thread of the GPU. Of course, it is essential that the three xor-like
+PRNGs used for our computation have different parameters. So we chose them
+randomly with another PRNG. As the initialisation is performed by the CPU, we
+have chosen to use the ISAAC PRNG [ref] to initalize all the parameters for the
+GPU version of our PRNG. The implementation of the three xor-like PRNGs is
straightforward as soon as their parameters have been allocated in the GPU
memory. Each xor-like PRNGs used works with an internal number $x$ which keeps
the last generated random numbers. Other internal variables are also used by the
All the tests performed to pass the BigCrush of TestU01 succeeded. Different
number of threads have been tested upto $10$ millions.
+\begin{remark}
+Algorithm~\ref{algo:gpu_kernel} has the advantage to manipulate independent
+PRNGs, so this version is easily usable on a cluster of computer. The only thing
+to ensure is to use a single ISAAC PRNG. For this, a simple solution consists in
+using a master node for the initialization which computes the initial parameters
+for all the differents nodes involves in the computation.
+\end{remark}
+
+\subsection{Version more suited to GPU}
+
+As GPU offers shared memory mechanism between threads of the same block, it is
+possible to use this in order to simplify the previous algorithm, i.e. using
+less than 3 xor-like PRNGs. The solution consists in
+
+ threads of the same block compute a random
+number and uses other random numbers of
+
\section{Experiments}
Differents experiments have been performed in order to measure the generation speed.
