doc/html/taskplugins.shtml

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

<h1><a name="top">SLURM Task Plugin Programmer Guide</a></h1>

<h2> Overview</h2>
<p> This document describes SLURM task management plugins and the API
that defines them. It is intended as a resource to programmers wishing
to write their own SLURM scheduler plugins. This is version 1 of the API.</p>

<p>SLURM task management plugins are SLURM plugins that implement the
SLURM task management API described herein. They would typically be
used to control task affinity (i.e. binding tasks to processors).
They must conform to the SLURM Plugin API with the following
specifications:</p>
<p><span class="commandline">const char plugin_type[]</span><br>
The major type must be &quot;task.&quot; The minor type can be any recognizable
abbreviation for the type of task management. We recommend, for example:</p>
<ul>
<li><b>affinity</b>&#151;A plugin that implements task binding to processors.
The actual mechanism used to task binding is dependent upon the available
infrastructure as determined by the "configure" program when SLURM is built
and the value of the <b>TaskPluginParam</b> as defined in the <b>slurm.conf</b>
(SLURM configuration file).</li>
<li><b>none</b>&#151;A plugin that implements the API without providing any
services. This is the default behavior and provides no task binding.</li>
</ul>

<p>The <span class="commandline">plugin_name</span> and
<span class="commandline">plugin_version</span>
symbols required by the SLURM Plugin API require no specialization
for task support.
Note carefully, however, the versioning discussion below.</p>

<p class="footer"><a href="#top">top</a></p>

<h2>Data Objects</h2>
<p>The implementation must maintain (though not necessarily directly export) an
enumerated <span class="commandline">errno</span>  to allow SLURM to discover
as practically as possible the reason for any failed API call.
These values must not be used as return values in integer-valued functions
in the API. The proper error return value from integer-valued functions is
SLURM_ERROR.</p>
<p class="footer"><a href="#top">top</a></p>

<h2>API Functions</h2>
<p>The following functions must appear. Functions which are not implemented should
be stubbed.</p>

<p class="commandline">int task_slurmd_batch_request (uint32_t job_id,
batch_job_launch_msg_t *req);</p>
<p style="margin-left:.2in"><b>Description</b>: Prepare to launch a batch job.
Establish node, socket, and core resource availability for it.
Executed by the <b>slurmd</b> daemon as user root.</p>
<p style="margin-left:.2in"><b>Arguments</b>:<br>
<span class="commandline">job_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the job to be started.<br>
<span class="commandline">req</span>&nbsp;&nbsp;&nbsp;(input/output)
Batch job launch request specification.
See <b>src/common/slurm_protocol_defs.h</b> for the
data structure definition.</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<p class="commandline">int task_slurmd_launch_request (uint32_t job_id,
launch_tasks_request_msg_t *req, uint32_t node_id);</p>
<p style="margin-left:.2in"><b>Description</b>: Prepare to launch a job.
Establish node, socket, and core resource availability for it.
Executed by the <b>slurmd</b> daemon as user root.</p>
<p style="margin-left:.2in"><b>Arguments</b>:<br>
<span class="commandline">job_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the job to be started.<br>
<span class="commandline">req</span>&nbsp;&nbsp;&nbsp;(input/output)
Task launch request specification including node, socket, and
core specifications.
See <b>src/common/slurm_protocol_defs.h</b> for the
data structure definition.<br>
<span class="commandline">node_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the node on which resources are being acquired (zero origin).</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<p class="commandline">int task_slurmd_reserve_resources (uint32_t job_id,
launch_tasks_request_msg_t *req, uint32_t node_id);</p>
<p style="margin-left:.2in"><b>Description</b>: Reserve resources for
the initiation of a job. Executed by the <b>slurmd</b> daemon as user root.</p>
<p style="margin-left:.2in"><b>Arguments</b>:<br>
<span class="commandline">job_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the job being started.<br>
<span class="commandline">req</span>&nbsp;&nbsp;&nbsp;(input)
Task launch request specification including node, socket, and
core specifications.
See <b>src/common/slurm_protocol_defs.h</b> for the
data structure definition.<br>
<span class="commandline">node_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the node on which the resources are being acquired
(zero origin).</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<p class="commandline">int task_slurmd_suspend_job (uint32_t job_id);</p>
<p style="margin-left:.2in"><b>Description</b>: Temporarily release resources
previously reserved for a job.
Executed by the <b>slurmd</b> daemon as user root.</p>
<p style="margin-left:.2in"><b>Arguments</b>:
<span class="commandline">job_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the job which is being suspended.</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<p class="commandline">int task_slurmd_resume_job (uint32_t job_id);</p>
<p style="margin-left:.2in"><b>Description</b>: Reclaim resources which
were previously released using the task_slurmd_suspend_job function.
Executed by the <b>slurmd</b> daemon as user root.</p>
<p style="margin-left:.2in"><b>Arguments</b>:
<span class="commandline">job_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the job which is being resumed.</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<p class="commandline">int task_slurmd_release_resources (uint32_t job_id);</p>
<p style="margin-left:.2in"><b>Description</b>: Release resources previously
reserved for a job. Executed by the <b>slurmd</b> daemon as user root.</p>
<p style="margin-left:.2in"><b>Arguments</b>:
<span class="commandline">job_id</span>&nbsp;&nbsp;&nbsp;(input)
ID of the job which has completed.</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<p class="commandline">int task_pre_setuid (slurmd_job_t *job);</p>
<p style="margin-left:.2in"><b>Description</b>: task_pre_setuid() is called
before setting the UID for the user to launch his jobs.
Executed by the <b>slurmstepd</b> program as user root.</p>
<p style="margin-left:.2in"><b>Arguments</b>:
<span class="commandline">job</span>&nbsp;&nbsp;&nbsp;(input)
pointer to the job to be initiated.
See <b>src/slurmd/slurmstepd/slurmstepd_job.h</b> for the
data structure definition.</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<p class="commandline">int task_pre_launch (slurmd_job_t *job);</p>
<p style="margin-left:.2in"><b>Description</b>: task_pre_launch() is called
prior to exec of application task.
Executed by the <b>slurmstepd</b> program as the job's owner.
It is followed by <b>TaskProlog</b> program (as configured in <b>slurm.conf</b>)
and <b>--task-prolog</b> (from <b>srun</b> command line).</p>
<p style="margin-left:.2in"><b>Arguments</b>:
<span class="commandline">job</span>&nbsp;&nbsp;&nbsp;(input)
pointer to the job to be initiated.
See <b>src/slurmd/slurmstepd/slurmstepd_job.h</b> for the
data structure definition.</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<a name="get_errno"><p class="commandline">int task_post_term (slurmd_job_t *job);</p></a>
<p style="margin-left:.2in"><b>Description</b>: task_term() is called
after termination of job step.
Executed by the <b>slurmstepd</b> program as the job's owner.
It is preceded by <b>--task-epilog</b> (from <b>srun</b> command line)
followed by <b>TaskEpilog</b> program (as configured in <b>slurm.conf</b>).</p>
<p style="margin-left:.2in"><b>Arguments</b>:
<span class="commandline">job</span>&nbsp;&nbsp;&nbsp;(input)
pointer to the job which has terminated.
See <b>src/slurmd/slurmstepd/slurmstepd_job.h</b> for the
data structure definition.</p>
<p style="margin-left:.2in"><b>Returns</b>: SLURM_SUCCESS if successful.
On failure, the plugin should return SLURM_ERROR and set the errno to an
appropriate value to indicate the reason for failure.</p>

<h2>Versioning</h2>
<p> This document describes version 1 of the SLURM Task Plugin API.
Future releases of SLURM may revise this API.</p>
<p class="footer"><a href="#top">top</a></p>

<p style="text-align:center;">Last modified 19 February 2009</p>

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