Skip to content

Commit

Permalink
Deployed af8da95 with MkDocs version: 1.6.1
Browse files Browse the repository at this point in the history
  • Loading branch information
Unknown committed Nov 12, 2024
1 parent 7b71d74 commit 41c050e
Show file tree
Hide file tree
Showing 4 changed files with 86 additions and 50 deletions.
2 changes: 1 addition & 1 deletion search/search_index.json

Large diffs are not rendered by default.

84 changes: 42 additions & 42 deletions sitemap.xml
Original file line number Diff line number Diff line change
Expand Up @@ -2,170 +2,170 @@
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/information/download/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/information/faq/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/information/notes/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/information/troubleshooting/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/information/tutorial/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/introduction/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/introduction/register/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/introduction/updates/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/basic/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/job/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/node/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/others/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/partition/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/qos/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/sacct/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/salloc/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/sattach/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/sbatch/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/sbcast/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/sinfo/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/squeue/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/srun/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/slurm/submission/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/compile-install/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/login/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/partition/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/quick-start/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/scow/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/abaqus/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/amber/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/comsol/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/cp2k/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/gaussian/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/gromacs/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/lammps/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/matlab/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/orca/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/pytorch/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
<url>
<loc>https://ai4ec.ikkem.com/ikkem-hpc/doc/usage/apps/vasp/</loc>
<lastmod>2024-11-11</lastmod>
<lastmod>2024-11-12</lastmod>
</url>
</urlset>
Binary file modified sitemap.xml.gz
Binary file not shown.
50 changes: 43 additions & 7 deletions usage/apps/cp2k/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -615,9 +615,9 @@
<ul class="md-nav__list">

<li class="md-nav__item">
<a href="#_2" class="md-nav__link">
<a href="#oom" class="md-nav__link">
<span class="md-ellipsis">
内存泄漏
内存 OOM 问题排查
</span>
</a>

Expand Down Expand Up @@ -1386,9 +1386,9 @@
<ul class="md-nav__list">

<li class="md-nav__item">
<a href="#_2" class="md-nav__link">
<a href="#oom" class="md-nav__link">
<span class="md-ellipsis">
内存泄漏
内存 OOM 问题排查
</span>
</a>

Expand Down Expand Up @@ -1555,13 +1555,49 @@ <h2 id="cp2k_1">嘉庚智算上的CP2K<a class="headerlink" href="#cp2k_1" title
</div>
</div>
<h2 id="_1">已知问题<a class="headerlink" href="#_1" title="Permanent link">&para;</a></h2>
<h3 id="_2">内存泄漏<a class="headerlink" href="#_2" title="Permanent link">&para;</a></h3>
<p>一些版本可能存在严重的内存泄漏 Issue,若遇到此问题,建议用户关闭 ELPA 功能,使用 Scalapack 进行对角化。</p>
<h3 id="oom">内存 OOM 问题排查<a class="headerlink" href="#oom" title="Permanent link">&para;</a></h3>
<p>请首先检查所申请内存是否足够,若所运行计算可以跑满整个节点,请设置 <code>#SBATCH --mem=251G</code></p>
<p>一些版本可能存在严重的内存泄漏 Issue,若遇到内存 OOM 问题,建议用户关闭 ELPA 功能(特别是使用 DIIS 做对角化的情况),使用 Scalapack 进行对角化。</p>
<div class="highlight"><pre><span></span><code>&amp;GLOBAL
PREFERRED_DIAG_LIBRARY SL
&amp;END GLOBAL
</code></pre></div>
<p>若仍然存在问题,建议尝试降低每个节点上的核数,即调整 <code>--ntasks-per-node</code> 为更低的值。</p>
<p>若仍然存在问题,请进一步按照以下步骤进行测试:</p>
<ol>
<li>
<p>若为多节点并行任务,请调查任务是否正确并行在每个节点上,当使用 <code>mpirun</code> 时,一般需要确保 <code>-np</code> 的值为所有节点的进程总数(通常使用 <code>popt</code> 或者 <code>OMP_NUM_THREADS=1</code> 时为核数)</p>
</li>
<li>
<p>检查输入参数中交换关联泛函部分的 <code>MAX_MEMORY</code> 设置(单位为MB),若该值太大则需要根据节点总量适当缩减。
例如采用杂化泛函进行模拟时,</p>
<div class="highlight"><pre><span></span><code>&amp;XC
&amp;HF
&amp;MEMORY
MAX_MEMORY 1500
EPS_STORAGE_SCALING 0.1
&amp;END
&amp;END
&amp;END XC
</code></pre></div>
</li>
</ol>
<p>具体数值请根据体系情况进行测试,确保不会造成OOM</p>
<ol>
<li>
<p>使用psmp版本并尝试提高OMP线程数,以减少总进程数,线程间可以共享内存</p>
<p>例如总共申请64核,采用16个进程,每个进程4个线程:</p>
<div class="highlight"><pre><span></span><code>...
<span class="c1">#SBATCH -N 1</span>
<span class="c1">#SBATCH --ntasks-per-node=64 </span>
...
<span class="nb">export</span><span class="w"> </span><span class="nv">OMP_NUM_THREADS</span><span class="o">=</span><span class="m">4</span>
mpirun<span class="w"> </span>-np<span class="w"> </span><span class="m">16</span><span class="w"> </span>cp2k.psmp<span class="w"> </span>-i<span class="w"> </span>input
</code></pre></div>
</li>
<li>
<p>若问题仍然存在或者出现了意料之外的报错(此时请恢复到 <code>popt</code> 版本),建议尝试降低每个节点上的核数,即调整 <code>#SBATCH --ntasks-per-node</code> 为更低的值。</p>
</li>
</ol>



Expand Down

0 comments on commit 41c050e

Please sign in to comment.