build based on 4f0746e

This commit is contained in:
Documenter.jl committed 2024-08-20 13:33:00 +00:00
1 parent cd80d19161
commit e8001495c2
25 files changed
+382 -416

No files matched your search

+137 -173
View File
@@ -7584,6 +7584,7 @@ a.anchor-link {
<span class="n">jacobi_1_check</span><span class="p">(</span><span class="n">answer</span><span class="p">)</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">answer_checker</span><span class="p">(</span><span class="n">answer</span><span class="p">,</span><span class="w"> </span><span class="s">"d"</span><span class="p">)</span>
<span class="n">jacobi_2_check</span><span class="p">(</span><span class="n">answer</span><span class="p">)</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">answer_checker</span><span class="p">(</span><span class="n">answer</span><span class="p">,</span><span class="w"> </span><span class="s">"b"</span><span class="p">)</span>
<span class="n">jacobi_3_check</span><span class="p">(</span><span class="n">answer</span><span class="p">)</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">answer_checker</span><span class="p">(</span><span class="n">answer</span><span class="p">,</span><span class="w"> </span><span class="s">"c"</span><span class="p">)</span>
<span class="n">println</span><span class="p">(</span><span class="s">"🥳 Well done! "</span><span class="p">)</span>
</pre></div>
</div>
</div>
@@ -7684,7 +7685,7 @@ a.anchor-link {
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<div class="alert alert-block alert-info">
<b>Note:</b> In our version of the jacobi method, we return after a given number of iterations. Other stopping criteria are possible. For instance, iterate until the difference between u and u_new is below a tolerance.
<b>Note:</b> In our version of the jacobi method, we return after a given number of iterations. Other stopping criteria are possible. For instance, iterate until the difference between u and u_new is below a tolerance.
</div>
</div>
</div>
@@ -7827,7 +7828,7 @@ d) The inner, but not the outer</code></pre>
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<h3 id="Parallelization-strategy">Parallelization strategy<a class="anchor-link" href="#Parallelization-strategy"></a></h3><p>Remember that a sufficiently large grain size is needed to achieve performance in a distributed algorithm. For Jacobi, one could update each entry of vector <code>u_new</code> in a different worker, but this would not be efficient. Instead, we use a parallelization strategy with a larger grain size that is analogous to the algorithm 3 we studied for the matrix-matrix multiplication:</p>
<h3 id="Parallelization-strategy">Parallelization strategy<a class="anchor-link" href="#Parallelization-strategy"></a></h3><p>Remember that a sufficiently large grain size is needed to achieve performance in a distributed algorithm. For Jacobi, one could update each entry of vector <code>u_new</code> in a different process, but this would not be efficient. Instead, we use a parallelization strategy with a larger grain size that is analogous to the algorithm 3 we studied for the matrix-matrix multiplication:</p>
<ul>
<li>Each worker updates a consecutive section of the array <code>u_new</code></li>
</ul>
@@ -7849,7 +7850,7 @@ d) The inner, but not the outer</code></pre>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=3f90d701">
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=22dc6e54">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
@@ -7857,7 +7858,65 @@ d) The inner, but not the outer</code></pre>
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<h3 id="Data-dependencies">Data dependencies<a class="anchor-link" href="#Data-dependencies"></a></h3><p>Recall the Jacobi update:</p>
<p><code>u_new[i] = 0.5*(u[i-1]+u[i+1])</code></p>
<p>Thus, in order to update the local entries in <code>u_new</code>, we also need remote entries of vector <code>u</code> located in neighboring processes. Figure below shows the entries of <code>u</code> needed to update the local entries of <code>u_new</code> in a particular process (CPU 2).</p>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=ba4113af">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<p>Note that an entry in the interior of the locally stored vector can be updated using local data only. For this one, communication is not needed.</p>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=97a5079d">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<div>
<img align="left" alt="No description has been provided for this image" src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA6gAAADyCAYAAABAvOgkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAewgAAHsIBbtB1PgAAABl0RVh0U29mdHdhcmUAd3d3Lmlua3NjYXBlLm9yZ5vuPBoAABgzSURBVHic7d19mF11YSfw750kBBIigWJiADUEBOXdRYvgCwtCX/VhH0TbLRVrdUlod/Gpq0+7a1uoL1316RarqzSpZXe77q61Pmu73bZYEFOtFUSs4BsCQhCEEEyUd4aQ3P3jnGHOTO6duTOZmfObyefzPPfJ79x7zr1fZm5O7pfzO+d2Nm7c2A0AAAC0bKjtAAAAAJAoqAAAABRi8bjln+50OlvaCAIAAMC+pdvtrk3y2ZHlMQW10+lsufjii2+b61AAAADsezZt2pRud/SySKb4AgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAowuK2AwAAPb05yZIkVyf5fstZAGBOKKgAUKY/SrIiyWsz/YK6JMmJSU5N8pIkB9f3/26SW/c2YGH+KMnLk/z3JB9p4fUXp/p9JcnjSYZbyAAw7ymoALAw/U6SdyVZ2uOxD89xlrlwdKoifu0cvd6aJOemKv6nJjklybL6sfVJNs1RDoAFRUEFgDK9LdUR0Fumuf2aVOW0m+T2JPcmOXtmopHkgizMog/QKgUVAMr0X/dy+08m+VSSryV5OMlLk3xlb0MV7Oenud2FqaZBfzrJV6e47ZYkN9W3byf5y2lmAKCmoAJAmdamutr+1lTnNE7VF2Y0zcL1r1IdDf1uplZQP5qx57oun8lQAPsqXzMDAGW6Jcn3YlruoN6c5P1JfnaOXm/3HL0OwD7FEVQAYCG4IMnP1eO/azMIANPnCCoAAABFUFABAAAogoIKAABAEZyDCgDzx7IkZ/R57LYk35/DLPPJLyZ5S5/HTqr/fEeSX+qzzvokd850KAD2pKACwPxxWJJr+jz29iRXzGGW+eTIJOdMss5x9a2XA2c2DgD9KKgAMH88nuTaPo85etrfZ5Lc0eextyd5WZKPJdncZ527ZyETAD0oqAAwf9yX5Ny2Q8xDt9a3Xt5Q//nVJH8xN3EA6MdFkgAAACiCggoAAEARTPEFgIVpbZINjeU1jfGGJK9pLH88/c/RBIA5o6ACwML03CS/2eexC8ctXxMFdaoOTvK9xnKnMb4iyfsby7+VZNNchAKY7xRUACjTFUmWZvrF8Z4kHxhwXVepnbpOqpLay7L6NmL/2Y8DsDAoqABQpsv2cvstqY7c7SvekqoU/niK2/3bVEeaH5zidj9OctSA626f4nMD7LMUVABgIdg6ze0emOZ2u5PcOc1tAejDVXwBAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAirC47QDQy5VXXnn40NDQG9rOMZGrrrrqVUk6SXLXXXfdsm3bth+1HGm8E5McUo/vSXJni1l6OSLJUfX40SQ3tZhlEFcmebLtEAzkDUkObzvEBJ6fZG09fijJ19uL0tOzkry4sfzFJLtbytLPKzP6P9m/kWRHi1km87Uk/9B2CAby/CTntx1iAp0kr2osfz3VPqQkJydZWY/vTrKlvSh7Wrt27fNXr169NklWrlz50Pnnn1/a/neM+++//yOXX375023nmGsKKkVatGjR0d1u9w/bzjGRG2+8Mbt3P/OZ7bw2sywQF7YdYBJ/FgV1vrg0ycvbDjEFF7UdYBKl799Kz/ehKKjzxYuSFP3ZY5zS3/vF2bJlS7Zs2ZIkWbduXVL4/nfNmjUbk+xzBdUUXwAAAIrgCCrzxS1tB+jhpJHB0qVLtw4PD29rM0wPx2X07/gTSW5vMUsva1NNJUySbqppeiXZP8kxbYdgrz1Q30pyVJLl1XBod7KssOn3wwclO5/duOObKW+K70mjw/13JIsLm+L7xGHJrmVtp2CvlfbZo5Pq9J0R9yX5YUtZ+jk+yaJ6/FiS77WYZQ9Lly5dNzw8fGCSdLvd3an2byVZluTotkO0TUFlvnjJ+vXrd7YdomloaKg7Mj799NM/unnz5ve2maeHB5McWo+/leSlLWbp5a+TvKYeP5nqvJWSvDjVuWPMb3+c5PK2Q4yzOcmZ1XD5E8nDhU1vP+P1yZff0bjj5UkebitNH93R4Wn/J9n8J+1F6WXVFcmDZ7Sdgr2yK+X9u7QsVekb8YEkH24pSz8/TnJQPf7nVOeLF+OEE0647qabbjorSXbu3PnY+vXri/odb9q06Yxut/ultnO0zRRfAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAvHjUm+Wt+OG3CbzzS2+alZygUAADCQxW0HYMacmqRTj5cPuM0JSY6ux4fMeCIAAIApcAQVAACAIiioAAAAFEFBBQAAoAgKKgAAAEVQUAEAACiCggoAAEARFFQAAACKoKACAABQBAUVAACAIiio+7YlbQcAAAAYoaAuHMON8dIB1u8kOXSWsgAAAEyZgrpwPNwYP2eA9Y9PsnyWsgAAAEyZgrpw3N4Ynz7A+r86W0EAAACmQ0FdOG5ojC9M8qwJ1v3JJL8+u3EAAACmZnHbAZgxn0jy9nq8OsmfJ/nlJNsb63SS/OskH0t1gaThDHa+aus6nc5LN27cuKvtHE2XXHLJM+OtW7cekeS09tL01Pz7vTzl5Tu4MR5KefmOaTsAM+LwlPfeWjE63DWUvO349qL0sn38aSKnJnm8jSSD+eGzy/sZPnVg2wmYEaXtO8Z/Znt+ysu4qDE+MIXle+SRR57Z/3Y6nUUbN24sKl+32z2u7QwlUFAXjn9OVUp/oV7+mSR3Jdmc5P5UZeCMVB/WkuSyJG9McvScppymbrf7pbYzTOTWW29dn2R92zkm8KIk17cdYgJLU3Y+5q+31rdCPX5A8uGr2k4xievaDjCxb51X3WBGLUr5/y69PaMHJ0p0Sgr7Gd52223PjJcsWbIsheWjoqAuLBtS/d+0l9XLK5K8dtw63SQfTPKeVAUVAACgCArqwvLjJGcluTTVEYMXNB7bleTqJB9Kcm193z8l2VKPt85NRAAAgN4U1IXnyVRHSD+Y5Nmpzkd9PMm9SZ4at+6b5jba4O67774vrlu3ruhzeF73utcte/rppztJ8o1vfOPJO+64o6hzZFNNmx35O74ze/7+27Y4o+fT7Er13i3ZY20HYGDnZOx5UKVZkmS/elzie38oyQGN5RLf+8tSXVchqX5+pe1/m3a2HYCBXZPqvMmSNb8i8Ikku9sK0sf+Gd3/PpXC3v+nnHLKkiOPPHK/JFmxYsWuAw44oLT97xgXXXRRwef/z57Oxo0bu88sdDrHXnzxxbdNtAEAAADMhE2bNh3T7Xa/O7Lsa2YAAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgCz7VNJrktydttBACibggoAzLZXJDkryeq2gwBQNgUVAACAIixuOwAAsOCdkmRRkh+1HQSAsimoAMBs29Z2AADmBwUVAJhtb05yYJKrk9zechYACqagAgCz7X1J1iT5pSioAEzARZIAAAAogoIKAABAERRUAAAAiqCgAgAAUAQFFQAAgCIoqAAAABRBQQUAAKAICioAAABFUFABAAAogoIKAABAERRUAAAAiqCgAgAAUITFbQcAABa885IsTXJr20EAKJuCCgDMthvbDgDA/GCKLwAAAEVQUAEAACiCggoAAEARFFQAAACKoKACAABQBAUVAACAIiioAAAAFEFBBQAAoAiL2w4AvWzatOmUbrf7sbZzTOR973vf8bt37+4kyfbt27c88cQTj7adaZy1SQ6sxz9MsrW9KD0dmuQ59fjJJHe0mGUQP5Pk4bZDMJArk5zcdogJrKpvSfJ4kjtbzNLL8iRHNpa/laTbUpZ+jk/Sqcd3J3mkxSyT+WSSD7cdgoGcluSKtkNMoJPqvT/iriSPtZSlnyNT7UOS5MEkD7SYZQ8HHXTQqhUrVqxKkjVr1jz+1re+tbT97xjDw8NnXXrppcNt55hrCiqlOijJ6W2HmMi9996b3bt3jyye2GaWAazI2A+cpVmR5Nlth5jEkrYDMLATU/j+o2FFktVth5jEy9oOMIkT2g4wiRvaDsDADs782XckyUltB5jEiiTr2g7R9NBDD+Whhx5Kkuy3337F73+XLl26qO0MbTDFFwAAgCI4gsq80Ol0LtvdOFxZiPeMDFatWvW5bdu2bW4xSy//IcmyevyDJH/cYpZefjnJsfX46SS/12KWXtYk+bW2Q7DXPp/kurZDjPOWVFPwk2Q4yXvbi9LTaUleM7p4wlXJop2tpenp5vWNhetS/Z5L8sYkx7Qdgr2yO8llbYcYZ0mS320s/02S61vK0s+7kuxfj+9O8vEWs+xh9erVb37gLine truncated
</div>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=343e60c3">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<p>However, to update the entries on the boundary of the locally stored vector we need entries stored on other processors.</p>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=fce954fe">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<div>
<img align="left" alt="No description has been provided for this image" src="data:image/png;base64,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 truncated
</div>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=3f90d701">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<p>Thus, in order to update the local entries in <code>u_new</code>, we also need some remote entries of vector <code>u</code> located in neighboring processes. Figure below shows the entries of <code>u</code> needed to update the local entries of <code>u_new</code> in a particular process (CPU 2).</p>
</div>
</div>
</div>
@@ -7960,7 +8019,7 @@ d) The inner, but not the outer</code></pre>
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<h3 id="Code">Code<a class="anchor-link" href="#Code"></a></h3><p>Take a look at the implementation below and try to understand it. Note that we have used MPIClustermanagers and Distributed just to run the MPI code on the notebook. When running it on a cluster, MPIClustermanagers and Distributed are not needed.</p>
<h3 id="MPI-Code">MPI Code<a class="anchor-link" href="#MPI-Code"></a></h3><p>Take a look at the implementation below and try to understand it.</p>
</div>
</div>
</div>
@@ -7972,7 +8031,7 @@ d) The inner, but not the outer</code></pre>
<div class="jp-InputPrompt jp-InputArea-prompt">In [ ]:</div>
<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
<div class="cm-editor cm-s-jupyter">
<div class="highlight hl-julia"><pre><span></span><span class="p">]</span><span class="w"> </span><span class="n">add</span><span class="w"> </span><span class="n">MPI</span><span class="w"> </span><span class="n">MPIClusterManagers</span>
<div class="highlight hl-julia"><pre><span></span><span class="p">]</span><span class="w"> </span><span class="n">add</span><span class="w"> </span><span class="n">MPI</span>
</pre></div>
</div>
</div>
@@ -7986,42 +8045,12 @@ d) The inner, but not the outer</code></pre>
<div class="jp-InputPrompt jp-InputArea-prompt">In [ ]:</div>
<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
<div class="cm-editor cm-s-jupyter">
<div class="highlight hl-julia"><pre><span></span><span class="k">using</span><span class="w"> </span><span class="n">MPIClusterManagers</span><span class="w"> </span>
<span class="k">using</span><span class="w"> </span><span class="n">Distributed</span>
<div class="highlight hl-julia"><pre><span></span><span class="k">using</span><span class="w"> </span><span class="n">MPI</span>
</pre></div>
</div>
</div>
</div>
</div>
</div><div class="jp-Cell jp-CodeCell jp-Notebook-cell jp-mod-noOutputs" id="cell-id=e0d63c6b">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea">
<div class="jp-InputPrompt jp-InputArea-prompt">In [ ]:</div>
<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
<div class="cm-editor cm-s-jupyter">
<div class="highlight hl-julia"><pre><span></span><span class="k">if</span><span class="w"> </span><span class="n">procs</span><span class="p">()</span><span class="w"> </span><span class="o">==</span><span class="w"> </span><span class="n">workers</span><span class="p">()</span>
<span class="w"> </span><span class="n">nw</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">3</span>
<span class="w"> </span><span class="n">manager</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPIWorkerManager</span><span class="p">(</span><span class="n">nw</span><span class="p">)</span>
<span class="w"> </span><span class="n">addprocs</span><span class="p">(</span><span class="n">manager</span><span class="p">)</span>
<span class="k">end</span>
</pre></div>
</div>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=d7fb9177">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<p>First, we implement the function to be called on the MPI ranks.</p>
</div>
</div>
</div>
</div><div class="jp-Cell jp-CodeCell jp-Notebook-cell jp-mod-noOutputs" id="cell-id=d39c7bb2">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
@@ -8030,132 +8059,99 @@ d) The inner, but not the outer</code></pre>
<div class="jp-InputPrompt jp-InputArea-prompt">In [ ]:</div>
<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
<div class="cm-editor cm-s-jupyter">
<div class="highlight hl-julia"><pre><span></span><span class="nd">@everywhere</span><span class="w"> </span><span class="n">workers</span><span class="p">()</span><span class="w"> </span><span class="k">begin</span>
<div class="highlight hl-julia"><pre><span></span><span class="n">code</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="k">quote</span>
<span class="w"> </span><span class="k">using</span><span class="w"> </span><span class="n">MPI</span>
<span class="w"> </span><span class="n">comm</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Comm_dup</span><span class="p">(</span><span class="n">MPI</span><span class="o">.</span><span class="n">COMM_WORLD</span><span class="p">)</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Init</span><span class="p">()</span>
<span class="w"> </span><span class="k">function</span><span class="w"> </span><span class="n">jacobi_mpi</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">niters</span><span class="p">)</span>
<span class="w"> </span><span class="n">comm</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">COMM_WORLD</span>
<span class="w"> </span><span class="n">nranks</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Comm_size</span><span class="p">(</span><span class="n">comm</span><span class="p">)</span>
<span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Comm_rank</span><span class="p">(</span><span class="n">comm</span><span class="p">)</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">mod</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">nranks</span><span class="p">)</span><span class="w"> </span><span class="o">!=</span><span class="w"> </span><span class="mi">0</span>
<span class="w"> </span><span class="n">println</span><span class="p">(</span><span class="s">"n must be a multiple of nranks"</span><span class="p">)</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Abort</span><span class="p">(</span><span class="n">comm</span><span class="p">,</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">n_own</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">div</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">nranks</span><span class="p">)</span>
<span class="w"> </span><span class="n">u</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">zeros</span><span class="p">(</span><span class="n">n_own</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span>
<span class="w"> </span><span class="n">load</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">div</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">nranks</span><span class="p">)</span>
<span class="w"> </span><span class="n">u</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">zeros</span><span class="p">(</span><span class="n">load</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span>
<span class="w"> </span><span class="n">u</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="o">-</span><span class="mi">1</span>
<span class="w"> </span><span class="n">u</span><span class="p">[</span><span class="k">end</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">1</span>
<span class="w"> </span><span class="n">u_new</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">copy</span><span class="p">(</span><span class="n">u</span><span class="p">)</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">t</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">1</span><span class="o">:</span><span class="n">niters</span>
<span class="w"> </span><span class="n">reqs</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Request</span><span class="p">[]</span>
<span class="w"> </span><span class="c"># Communication</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">!=</span><span class="w"> </span><span class="mi">0</span>
<span class="w"> </span><span class="n">neig_rank</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">rank</span><span class="o">-</span><span class="mi">1</span>
<span class="w"> </span><span class="n">req</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Isend</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="mi">2</span><span class="o">:</span><span class="mi">2</span><span class="p">),</span><span class="n">comm</span><span class="p">,</span><span class="n">dest</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">,</span><span class="n">tag</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="w"> </span><span class="n">push!</span><span class="p">(</span><span class="n">reqs</span><span class="p">,</span><span class="n">req</span><span class="p">)</span>
<span class="w"> </span><span class="n">req</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Irecv!</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="mi">1</span><span class="o">:</span><span class="mi">1</span><span class="p">),</span><span class="n">comm</span><span class="p">,</span><span class="n">source</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">,</span><span class="n">tag</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="w"> </span><span class="n">push!</span><span class="p">(</span><span class="n">reqs</span><span class="p">,</span><span class="n">req</span><span class="p">)</span>
<span class="w"> </span><span class="n">s</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">2</span>
<span class="w"> </span><span class="n">r</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">1</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Sendrecv!</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">s</span><span class="o">:</span><span class="n">s</span><span class="p">),</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">r</span><span class="o">:</span><span class="n">r</span><span class="p">),</span><span class="n">comm</span><span class="p">;</span><span class="n">dest</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">,</span><span class="n">source</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">)</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">!=</span><span class="w"> </span><span class="p">(</span><span class="n">nranks</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="n">neig_rank</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">rank</span><span class="o">+</span><span class="mi">1</span>
<span class="w"> </span><span class="n">s</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">n_own</span><span class="o">+</span><span class="mi">1</span>
<span class="w"> </span><span class="n">r</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">n_own</span><span class="o">+</span><span class="mi">2</span>
<span class="w"> </span><span class="n">req</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Isend</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">s</span><span class="o">:</span><span class="n">s</span><span class="p">),</span><span class="n">comm</span><span class="p">,</span><span class="n">dest</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">,</span><span class="n">tag</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="w"> </span><span class="n">push!</span><span class="p">(</span><span class="n">reqs</span><span class="p">,</span><span class="n">req</span><span class="p">)</span>
<span class="w"> </span><span class="n">req</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Irecv!</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">r</span><span class="o">:</span><span class="n">r</span><span class="p">),</span><span class="n">comm</span><span class="p">,</span><span class="n">source</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">,</span><span class="n">tag</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="w"> </span><span class="n">push!</span><span class="p">(</span><span class="n">reqs</span><span class="p">,</span><span class="n">req</span><span class="p">)</span>
<span class="w"> </span><span class="n">s</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">load</span><span class="o">+</span><span class="mi">1</span>
<span class="w"> </span><span class="n">r</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">load</span><span class="o">+</span><span class="mi">2</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Sendrecv!</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">s</span><span class="o">:</span><span class="n">s</span><span class="p">),</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">r</span><span class="o">:</span><span class="n">r</span><span class="p">),</span><span class="n">comm</span><span class="p">;</span><span class="n">dest</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">,</span><span class="n">source</span><span class="o">=</span><span class="n">neig_rank</span><span class="p">)</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Waitall</span><span class="p">(</span><span class="n">reqs</span><span class="p">)</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">i</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">2</span><span class="o">:</span><span class="p">(</span><span class="n">n_own</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="c"># Local computation</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">i</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">2</span><span class="o">:</span><span class="p">(</span><span class="n">load</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="n">u_new</span><span class="p">[</span><span class="n">i</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mf">0.5</span><span class="o">*</span><span class="p">(</span><span class="n">u</span><span class="p">[</span><span class="n">i</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">+</span><span class="n">u</span><span class="p">[</span><span class="n">i</span><span class="o">+</span><span class="mi">1</span><span class="p">])</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">u</span><span class="p">,</span><span class="w"> </span><span class="n">u_new</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">u_new</span><span class="p">,</span><span class="w"> </span><span class="n">u</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="k">return</span><span class="w"> </span><span class="n">u</span>
<span class="w"> </span><span class="k">end</span>
<span class="k">end</span>
</pre></div>
</div>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=6eab32d0">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<p>In order to check the result, we will compare it against the serial implementation. To this end, we need to define the serial implementation also in the workers.</p>
</div>
</div>
</div>
</div><div class="jp-Cell jp-CodeCell jp-Notebook-cell jp-mod-noOutputs" id="cell-id=f1a8db8f">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea">
<div class="jp-InputPrompt jp-InputArea-prompt">In [ ]:</div>
<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
<div class="cm-editor cm-s-jupyter">
<div class="highlight hl-julia"><pre><span></span><span class="nd">@everywhere</span><span class="w"> </span><span class="n">workers</span><span class="p">()</span><span class="w"> </span><span class="k">function</span><span class="w"> </span><span class="n">jacobi</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">niters</span><span class="p">)</span>
<span class="w"> </span><span class="n">u</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">zeros</span><span class="p">(</span><span class="n">n</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span>
<span class="w"> </span><span class="n">u</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="o">-</span><span class="mi">1</span>
<span class="w"> </span><span class="n">u</span><span class="p">[</span><span class="k">end</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">1</span>
<span class="w"> </span><span class="n">u_new</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">copy</span><span class="p">(</span><span class="n">u</span><span class="p">)</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">t</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">1</span><span class="o">:</span><span class="n">niters</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">i</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">2</span><span class="o">:</span><span class="p">(</span><span class="n">n</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="n">u_new</span><span class="p">[</span><span class="n">i</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mf">0.5</span><span class="o">*</span><span class="p">(</span><span class="n">u</span><span class="p">[</span><span class="n">i</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">+</span><span class="n">u</span><span class="p">[</span><span class="n">i</span><span class="o">+</span><span class="mi">1</span><span class="p">])</span>
<span class="w"> </span><span class="c"># Gather the results</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">!=</span><span class="mi">0</span>
<span class="w"> </span><span class="n">lb</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">2</span>
<span class="w"> </span><span class="n">ub</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">load</span><span class="o">+</span><span class="mi">1</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Send</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">lb</span><span class="o">:</span><span class="n">ub</span><span class="p">),</span><span class="n">comm</span><span class="p">,</span><span class="n">dest</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="w"> </span><span class="n">u_all</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">zeros</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="w"> </span><span class="c"># This will nevel be used</span>
<span class="w"> </span><span class="k">else</span>
<span class="w"> </span><span class="n">u_all</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">zeros</span><span class="p">(</span><span class="n">n</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span>
<span class="w"> </span><span class="c"># Set boundary</span>
<span class="w"> </span><span class="n">u_all</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="o">-</span><span class="mi">1</span>
<span class="w"> </span><span class="n">u_all</span><span class="p">[</span><span class="k">end</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">1</span>
<span class="w"> </span><span class="c"># Set data for rank 0</span>
<span class="w"> </span><span class="n">lb</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">2</span>
<span class="w"> </span><span class="n">ub</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">load</span><span class="o">+</span><span class="mi">1</span>
<span class="w"> </span><span class="n">u_all</span><span class="p">[</span><span class="n">lb</span><span class="o">:</span><span class="n">ub</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="n">lb</span><span class="o">:</span><span class="n">ub</span><span class="p">)</span>
<span class="w"> </span><span class="c"># Set data for other ranks</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">other_rank</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">1</span><span class="o">:</span><span class="p">(</span><span class="n">nranks</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="n">lb</span><span class="w"> </span><span class="o">+=</span><span class="w"> </span><span class="n">load</span>
<span class="w"> </span><span class="n">ub</span><span class="w"> </span><span class="o">+=</span><span class="w"> </span><span class="n">load</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Recv!</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u_all</span><span class="p">,</span><span class="n">lb</span><span class="o">:</span><span class="n">ub</span><span class="p">),</span><span class="n">comm</span><span class="p">;</span><span class="n">source</span><span class="o">=</span><span class="n">other_rank</span><span class="p">)</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">u</span><span class="p">,</span><span class="w"> </span><span class="n">u_new</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">u_new</span><span class="p">,</span><span class="w"> </span><span class="n">u</span>
<span class="w"> </span><span class="k">return</span><span class="w"> </span><span class="n">u_all</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">u</span>
<span class="k">end</span>
</pre></div>
</div>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=d2b04c67">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<p>Finally, we call the parallel function on the workers, gather the results on the root rank, and compare against the sequential solution.</p>
</div>
</div>
</div>
</div><div class="jp-Cell jp-CodeCell jp-Notebook-cell jp-mod-noOutputs" id="cell-id=68851107">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea">
<div class="jp-InputPrompt jp-InputArea-prompt">In [ ]:</div>
<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
<div class="cm-editor cm-s-jupyter">
<div class="highlight hl-julia"><pre><span></span><span class="nd">@everywhere</span><span class="w"> </span><span class="n">workers</span><span class="p">()</span><span class="w"> </span><span class="k">begin</span>
<span class="w"> </span><span class="c"># Call jacobi in parallel</span>
<span class="w"> </span><span class="n">niters</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">10</span>
<span class="w"> </span><span class="n">load</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">4</span>
<span class="w"> </span><span class="n">nranks</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Comm_size</span><span class="p">(</span><span class="n">comm</span><span class="p">)</span>
<span class="w"> </span><span class="n">n</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">load</span><span class="o">*</span><span class="n">nranks</span>
<span class="w"> </span><span class="n">u</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">jacobi_mpi</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">niters</span><span class="p">)</span>
<span class="w"> </span><span class="c"># Gather results in root process and check</span>
<span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Comm_rank</span><span class="p">(</span><span class="n">comm</span><span class="p">)</span>
<span class="w"> </span><span class="n">n_own</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">div</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">nranks</span><span class="p">)</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">==</span><span class="w"> </span><span class="mi">0</span>
<span class="w"> </span><span class="n">results</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">zeros</span><span class="p">(</span><span class="n">n</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span>
<span class="w"> </span><span class="n">results</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="o">-</span><span class="mi">1</span>
<span class="w"> </span><span class="n">results</span><span class="p">[</span><span class="n">n</span><span class="o">+</span><span class="mi">2</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">1</span>
<span class="w"> </span><span class="n">rcv</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">view</span><span class="p">(</span><span class="n">results</span><span class="p">,</span><span class="w"> </span><span class="mi">2</span><span class="o">:</span><span class="n">n</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="k">else</span>
<span class="w"> </span><span class="n">rcv</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="nb">nothing</span>
<span class="w"> </span><span class="k">function</span><span class="w"> </span><span class="n">jacobi</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">niters</span><span class="p">)</span>
<span class="w"> </span><span class="n">u</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">zeros</span><span class="p">(</span><span class="n">n</span><span class="o">+</span><span class="mi">2</span><span class="p">)</span>
<span class="w"> </span><span class="n">u</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="o">-</span><span class="mi">1</span>
<span class="w"> </span><span class="n">u</span><span class="p">[</span><span class="k">end</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">1</span>
<span class="w"> </span><span class="n">u_new</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">copy</span><span class="p">(</span><span class="n">u</span><span class="p">)</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">t</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">1</span><span class="o">:</span><span class="n">niters</span>
<span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">i</span><span class="w"> </span><span class="k">in</span><span class="w"> </span><span class="mi">2</span><span class="o">:</span><span class="p">(</span><span class="n">n</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span>
<span class="w"> </span><span class="n">u_new</span><span class="p">[</span><span class="n">i</span><span class="p">]</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mf">0.5</span><span class="o">*</span><span class="p">(</span><span class="n">u</span><span class="p">[</span><span class="n">i</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">+</span><span class="n">u</span><span class="p">[</span><span class="n">i</span><span class="o">+</span><span class="mi">1</span><span class="p">])</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">u</span><span class="p">,</span><span class="w"> </span><span class="n">u_new</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">u_new</span><span class="p">,</span><span class="w"> </span><span class="n">u</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">u</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Gather!</span><span class="p">(</span><span class="n">view</span><span class="p">(</span><span class="n">u</span><span class="p">,</span><span class="mi">2</span><span class="o">:</span><span class="n">n_own</span><span class="o">+</span><span class="mi">1</span><span class="p">),</span><span class="n">rcv</span><span class="p">,</span><span class="n">comm</span><span class="p">;</span><span class="n">root</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">==</span><span class="w"> </span><span class="mi">0</span>
<span class="w"> </span><span class="nd">@show</span><span class="w"> </span><span class="n">results</span><span class="w"> </span><span class="o">≈</span><span class="w"> </span><span class="n">jacobi</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">niters</span><span class="p">)</span>
<span class="w"> </span><span class="k">end</span><span class="w"> </span>
<span class="w"> </span><span class="k">function</span><span class="w"> </span><span class="n">testit</span><span class="p">(</span><span class="n">load</span><span class="p">)</span>
<span class="w"> </span><span class="n">comm</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">COMM_WORLD</span>
<span class="w"> </span><span class="n">nranks</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Comm_size</span><span class="p">(</span><span class="n">comm</span><span class="p">)</span>
<span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">MPI</span><span class="o">.</span><span class="n">Comm_rank</span><span class="p">(</span><span class="n">comm</span><span class="p">)</span>
<span class="w"> </span><span class="n">n</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">load</span><span class="o">*</span><span class="n">nranks</span>
<span class="w"> </span><span class="n">niters</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="mi">100</span>
<span class="w"> </span><span class="n">u_par</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">jacobi_mpi</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">niters</span><span class="p">)</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">rank</span><span class="w"> </span><span class="o">==</span><span class="w"> </span><span class="mi">0</span>
<span class="w"> </span><span class="c"># Compare agains serial</span>
<span class="w"> </span><span class="n">u_seq</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="n">jacobi</span><span class="p">(</span><span class="n">n</span><span class="p">,</span><span class="n">niters</span><span class="p">)</span>
<span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">u_par</span><span class="w"> </span><span class="o">≈</span><span class="w"> </span><span class="n">u_seq</span>
<span class="w"> </span><span class="n">println</span><span class="p">(</span><span class="s">"Test passed 🥳"</span><span class="p">)</span>
<span class="w"> </span><span class="k">else</span>
<span class="w"> </span><span class="n">println</span><span class="p">(</span><span class="s">"Test failed"</span><span class="p">)</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="k">end</span>
<span class="w"> </span><span class="n">testit</span><span class="p">(</span><span class="mi">3</span><span class="p">)</span>
<span class="k">end</span>
<span class="n">run</span><span class="p">(</span><span class="sb">`</span><span class="si">$</span><span class="p">(</span><span class="n">mpiexec</span><span class="p">())</span><span class="sb"> -np 4 julia --project=. -e </span><span class="si">$code</span><span class="sb">`</span><span class="p">);</span>
</pre></div>
</div>
</div>
@@ -8169,7 +8165,7 @@ d) The inner, but not the outer</code></pre>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<div class="alert alert-block alert-success">
<b>Question:</b> In function jacobi_mpi, how many messages per iteration are sent from a process away from the boundary?
<b>Question:</b> In function jacobi_mpi, how many messages per iteration are sent from a process away from the boundary (excluding the messages to collect the data on rank 0)?
</div>
<pre><code>a) 1
b) 2
@@ -8194,38 +8190,6 @@ d) 4</code></pre>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=075dd6d8">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea"><div class="jp-InputPrompt jp-InputArea-prompt">
</div><div class="jp-RenderedHTMLCommon jp-RenderedMarkdown jp-MarkdownOutput" data-mime-type="text/markdown">
<div class="alert alert-block alert-success">
<b>Question:</b> At the end of function jacobi_mpi ...
</div>
<pre><code>a) each process holds the complete solution.
b) the complete solution is gathered in the root process.
c) each process contains the solution for the local partition.
d) the ghost cells of u contain the initial values -1 and 1 in all processes.</code></pre>
</div>
</div>
</div>
</div><div class="jp-Cell jp-CodeCell jp-Notebook-cell jp-mod-noOutputs" id="cell-id=c3b58002">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
</div>
<div class="jp-InputArea jp-Cell-inputArea">
<div class="jp-InputPrompt jp-InputArea-prompt">In [ ]:</div>
<div class="jp-CodeMirrorEditor jp-Editor jp-InputArea-editor" data-type="inline">
<div class="cm-editor cm-s-jupyter">
<div class="highlight hl-julia"><pre><span></span><span class="n">answer</span><span class="w"> </span><span class="o">=</span><span class="w"> </span><span class="s">"x"</span><span class="w"> </span><span class="c"># replace x with a, b, c or d</span>
<span class="n">jacobi_3_check</span><span class="p">(</span><span class="n">answer</span><span class="p">)</span>
</pre></div>
</div>
</div>
</div>
</div>
</div>
<div class="jp-Cell jp-MarkdownCell jp-Notebook-cell" id="cell-id=c9aa2901">
<div class="jp-Cell-inputWrapper" tabindex="0">
<div class="jp-Collapser jp-InputCollapser jp-Cell-inputCollapser">
@@ -8446,7 +8410,7 @@ d) the ghost cells of u contain the initial values -1 and 1 in all processes.</c
<li>We update $N^2/P$ items per iteration</li>
<li>We need data from 2 neighbors (2 messages per iteration)</li>
<li>We communicate $N$ items per message</li>
<li>Communication/computation ratio is $O(P/N)$</li>
<li>Communication/computation ratio is $2N/(N^2/P) = 2P/N =O(P/N)$</li>
</ul>
</div>
</div>
@@ -8486,7 +8450,7 @@ d) the ghost cells of u contain the initial values -1 and 1 in all processes.</c
<li>We update $N^2/P$ items per iteration</li>
<li>We need data from 4 neighbors (4 messages per iteration)</li>
<li>We communicate $N/\sqrt{P}$ items per message</li>
<li>Communication/computation ratio is $O(\sqrt{P}/N)$</li>
<li>Communication/computation ratio is $ (4N/\sqrt{P})/(N^2/P)= 4\sqrt{P}/N =O(\sqrt{P}/N)$</li>
</ul>
</div>
</div>
@@ -8586,7 +8550,7 @@ d) the ghost cells of u contain the initial values -1 and 1 in all processes.</c
<li>Both 1d and 2d block partitions are potentially scalable if $P&lt;&lt;N$</li>
<li>The 2d block partition has the lowest communication complexity</li>
<li>The 1d block partition requires to send less messages (It can be useful if the fixed cost of sending a message is high)</li>
<li>The best strategy for a given problem size will thus depend on the machine, but the 2d block partition is the one used in practice since it has the lowest communication complexity.</li>
<li>The best strategy for a given problem size will thus depend on the machine.</li>
<li>Cyclic partitions are impractical for this application (but they are useful in others)</li>
</ul>
</div>