<!--#include virtual="header.txt"-->

<h1>Download Slurm</h1>
<p>Slurm source can be downloaded from
<a href="https://www.schedmd.com/download-slurm/">
https://www.schedmd.com/download-slurm/</a>.</p>

<h2>Related Software</h2>
<p>Note that the following software is not written or maintained by
SchedMD. Some of the software is required for certain functionality
(e.g. MySQL or MariaDB are required to use slurmdbd) while other software
was written to provide additional functionality for users or
administrators.</p>
<ul>
<li><b>Authentication</b> plugins identifies the user originating
a message.</li>
<ul>
<li><b>MUNGE</b> (recommended)<br>
In order to compile the "auth/munge" authentication plugin for Slurm,
you will need to build and install MUNGE, available from
<a href="https://dun.github.io/munge/">https://dun.github.io/munge/</a> and
<a href="http://packages.debian.org/src:munge">Debian</a> and
<a href="http://fedoraproject.org/">Fedora</a> and
<a href="http://packages.ubuntu.com/src:munge">Ubuntu</a>.</li>
</ul>

<li><b>Authentication</b> tools for users that work with Slurm.</li>
<ul>
<li><a href="https://github.com/hautreux/auks">AUKS</a><br>
AUKS is an utility designed to ease Kerberos V credential support addition
to non-interactive applications, like batch systems (Slurm, LSF, Torque, etc.).
It includes a plugin for the Slurm workload manager. AUKS is not used as
an authentication plugin by the Slurm code itself, but provides a mechanism
for the application to manage Kerberos V credentials.</li>
</ul>

<li><b>Databases</b> can be used to store accounting information.
See our <a href="accounting.html">Accounting</a> web page for more information.
</li>
<ul>
<li><a href="http://www.mysql.com/">MySQL</a></li>
<li><a href="https://mariadb.org/">MariaDB</a></li>
</ul><br>

<li><b>DRMAA (Distributed Resource Management Application API)</b><br>
<a href="https://github.com/psnc-apps/slurm-drmaa">PSNC DRMAA</a> for Slurm
is an implementation of <a href="http://www.gridforum.org/">Open Grid Forum</a>
<a href="http://www.drmaa.org/">DRMAA 1.0</a> (Distributed Resource Management Application API)
<a href="http://www.ogf.org/documents/GFD.133.pdf">specification</a> for submission
and control of jobs to <a href="https://slurm.schedmd.com">Slurm</a>.
Using DRMAA, grid applications builders, portal developers and ISVs can use
the same high-level API to link their software with different cluster/resource
management systems.<br><br>
There is a variant of PSNC DRMAA providing support for Slurm's --cluster option
available from
<a href="https://github.com/natefoo/slurm-drmaa">https://github.com/natefoo/slurm-drmaa</a>.<br><br>
Perl 6 DRMAA bindings are available from
<a href="https://github.com/scovit/Scheduler-DRMAA">https://github.com/scovit/Scheduler-DRMAA</a>.
</li><br>

<li><b>Hardware topology</b></li>
<ul>
<li><a href="http://www.open-mpi.org/projects/hwloc/">
Portable Hardware Locality (hwloc)</a></li>
<li><b>NOTE</b>: If you build Slurm or any MPI stack component with hwloc, note
that versions 2.5.0 through 2.7.0 (inclusive) of hwloc have a bug that pushes an
untouchable value into the environ array, causing a segfault when accessing it.
It is advisable to build with hwloc version 2.7.1 or later.</li>
<li>
Used by slurmd and PMIx client to get hardware topology information.
</li>
</ul>

<li><b>Hostlist</b><br>
A Python program used for manipulation of Slurm hostlists including
functions such as intersection and difference. Download the code from:<br>
<a href="http://www.nsc.liu.se/~kent/python-hostlist">
http://www.nsc.liu.se/~kent/python-hostlist</a><br><br>

Lua bindings for hostlist functions are also available here:<br>
<a href="https://github.com/grondo/lua-hostlist">
https://github.com/grondo/lua-hostlist</a><br>
<b>NOTE</b>: The Lua  hostlist functions do not support the bracketed numeric
ranges anywhere except at the end of the name (i.e. "tux[0001-0100]"
and "rack[0-3]_blade[0-63]" are not supported).</li><br>

<li><b>MPI</b> versions supported</li>
<ul>
<li><a href="https://software.intel.com/en-us/intel-mpi-library">Intel MPI</a></li>
<li><a href="https://www.mpich.org/">MPICH (a.k.a. MPICH2 / MPICH2)</a></li>
<li><a href="http://mvapich.cse.ohio-state.edu/">MVAPICH (a.k.a MVAPICH2)</a></li>
<li><a href="https://www.open-mpi.org">Open MPI</a></li>
</ul>

<li><b>Command wrappers</b><br>
There is a wrapper for Maui/Moab's showq command
<a href="https://github.com/pedmon/slurm_showq">here</a>.</li><br>

<li><b>Scripting interfaces</b></li>
<ul>
<li>A <b>Perl</b> interface is included in the Slurm distribution in the
<i>contribs/perlapi</i> directory and packaged in the <i>perlapi</i> RPM.</li>

<li><a href="https://github.com/pyslurm/">PySlurm</a> is a
Python/Cython module to interface with Slurm.
There is also a Python module to expand and collect hostlist expressions
available <a href="http://www.nsc.liu.se/~kent/python-hostlist/">
here</a>.</li>
</ul><br>

<li><b>SPANK Plugins</b><br>
SPANK provides a very generic interface for stackable plug-ins which
may be used to dynamically modify the job launch code in Slurm. SPANK
plugins may be built without access to Slurm source code. They need
only be compiled against Slurm&lsquo;s spank.h header file, added to the
SPANK config file plugstack.conf, and they will be loaded at runtime
during the next job launch. Thus, the SPANK infrastructure provides
administrators and other developers a low cost, low effort ability to
dynamically modify the runtime behavior of Slurm job launch.
Additional documentation can be found
<a href="https://slurm.schedmd.com/spank.html">here</a>.</li><br>

<li><b>Node Health Check</b><br>
Probably the most comprehensive and lightweight health check tool out
there is
<a href="https://github.com/mej/nhc">LBNL Node Health Check</a>.
It has integration with Slurm as well as Torque resource managers.</li><br>

<li><b>Accounting Tools</b>
<ul>
<li><b>UBMoD</b> is a web based tool for displaying accounting data from various
resource managers. It aggregates the accounting data from sacct into a MySQL
data warehouse and provide a front end web interface for browsing the data.
For more information, see the
<a href="http://ubmod.sourceforge.net/resource-manager-slurm.html">UDMod home page</a> and
<a href="https://github.com/ubccr/ubmod">source code</a>.<br></li>

<li><a href="http://xdmod.sourceforge.net"><b>XDMoD</b></a> (XD Metrics on Demand)
is an NSF-funded open source tool designed to audit and facilitate the utilization
of the XSEDE cyberinfrastructure by providing a wide range of metrics on XSEDE
resources, including resource utilization, resource performance, and impact on
scholarship and research.</li>
</ul></li><br>

<li><b>STUBL (Slurm Tools and UBiLities)</b><br>
STUBL is a collection of supplemental tools and utility scripts for Slurm.<br>
<a href="https://github.com/ubccr/stubl">STUBL home page</a>.</li><br>

<li><b>pestat</b><br>
Prints a consolidated compute node status line, with one line per node
including a list of jobs.<br>
<a href="https://github.com/OleHolmNielsen/Slurm_tools/tree/master/pestat">
Home page</a></li><br>
<!-- It would be nice to provide our own perl-based version of pestat -->

<li><b>Graphical Sdiag</b><br>
The sdiag utility is a diagnostic tool that maintains statistics on Slurm's
scheduling performance. You can run sdiag periodically or as you modify
Slurm's configuration. However if you want a historical view of these
statistics, you could save them in a time-series database and graph them over
time as performed with this tool:
<ul>
<li><a href="https://github.com/fasrc/slurm-diamond-collector">
  A collection of custom diamond collectors to gather various Slurm statistics</a></li>
<li><a href="https://github.com/collectd/collectd/pull/1198">Collectd</a>
  (for use with <a href="https://github.com/edf-hpc/jobmetrics">jobmetrics</a>)</li>
</ul></li><br>

<li><a id="json" href="https://github.com/json-c/json-c/wiki"><b>JSON</b></a>
	<p>Some Slurm plugins (<a href="rest.html">slurmrestd</a>,
	<a href="burst_buffer.html">burst_buffer/datawarp</a>,
	<a href="burst_buffer.html">burst_buffer/lua</a>,
	<a href="elasticsearch.html">jobcomp/elasticsearch</a>,
	<a href="jobcomp_kafka.html">jobcomp/kafka</a>
	and <a href="power_mgmt.html">power/cray_aries</a>) parse and/or
	serialize JSON format data. These plugins and slurmrestd are designed to
	make use of the JSON-C library for this purpose. Instructions for the
	build are as follows:</p>
	<pre>
git clone --depth 1 --single-branch -b json-c-0.15-20200726 https://github.com/json-c/json-c.git json-c
mkdir json-c-build
cd json-c-build
cmake ../json-c
make
sudo make install</pre>
	Declare the package configuration path before compiling Slurm
	(example provided for /bin/sh):
	<pre>
export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig/:$PKG_CONFIG_PATH</pre>
</li><br>

<li><a id="httpparser" href="https://github.com/nodejs/http-parser">
    <b>HTTP Parser</b></a>
	<p><a href="rest.html">slurmrestd</a> requires libhttp_parser
	(>=v2.6.0). Instructions for the build are as follows:</p>
	<pre>
git clone --depth 1 --single-branch -b v2.9.4 https://github.com/nodejs/http-parser.git http_parser
cd http_parser
make
sudo make install</pre>
	Add the following argument when running <i>configure</i> for Slurm:
	<pre>--with-http-parser=/usr/local/</pre>
</li><br>

<li><a id="yaml" href="https://github.com/yaml/libyaml">
    <b>YAML Parser</b></a>
	<p><a href="rest.html">slurmrestd</a> requires libyaml to support
	YAML. Instructions for the build are as follows:</p>
	<pre>
git clone --depth 1 --single-branch -b 0.2.5 https://github.com/yaml/libyaml libyaml
cd libyaml
./bootstrap
./configure
make
sudo make install</pre>
	Add the following argument when running <i>configure</i> for Slurm:
	<pre>--with-yaml=/usr/local/</pre>
</li><br>

<li><a id="jwt" href="https://github.com/benmcollins/libjwt">
    <b>JWT library</b></a>
	<p><a href="jwt.html">JWT authentication</a> requires libjwt.
	Instructions for the build are as follows:</p>
	<pre>
git clone --depth 1 --single-branch -b v1.12.0 https://github.com/benmcollins/libjwt.git libjwt
cd libjwt
autoreconf --force --install
./configure --prefix=/usr/local
make -j
sudo make install</pre>
	Add the following argument when running <i>configure</i> for Slurm:
	<pre>--with-jwt=/usr/local/</pre>
</li><br>

</ul>

<p style="text-align:center;">Last modified 12 March 2024</p>

<!--#include virtual="footer.txt"-->