diff --git a/doc/html/acct_gather_profile_plugins.shtml b/doc/html/acct_gather_profile_plugins.shtml
index 5ce48fac432908339b7417441f06c000593e0d01..a23bd9c75b53bdece25251a4d6f0ebd58a94290f 100644
--- a/doc/html/acct_gather_profile_plugins.shtml
+++ b/doc/html/acct_gather_profile_plugins.shtml
@@ -6,26 +6,26 @@
<h2> Overview</h2>
<p> This document describes SLURM profile accounting plugins and the API that
defines them. It is intended as a resource to programmers wishing to write
-their own SLURM profile accounting plugins.
+their own SLURM profile accounting plugins.
<p>A profiling plugin allows more detailed information on the execution of jobs
than can reasonably be kept in the accounting database. (All jobs may also not
be profiled.)
-<p>The plugin provides an API for making calls to store data at various
-points in a step's lifecycle. It collects data for <b>nodes</b>,
-<b>tasks</b> and periodic <b>samples</b>. The periodic samples are eventually
-consolidated into one <i>time series</i> dataset for each node of a job.
+<p>The plugin provides an API for making calls to store data at various
+points in a step's lifecycle. It collects data for <b>nodes</b>,
+<b>tasks</b> and periodic <b>samples</b>. The periodic samples are eventually
+consolidated into one <i>time series</i> dataset for each node of a job.
<p>The plugin's primary work is done within slurmstepd on the compute nodes.
It assumes a shared file system, presumably on the management network. This
-avoids having to transfer files back to the controller at step end. Data is
+avoids having to transfer files back to the controller at step end. Data is
typically gathered at job_acct_gather interval or acct_gather_energy interval
and the volume is not expected to be burdensome.
-<p>The reference implementation <i>(io_energy)</i> records I/O counts from the
-network interface (Infiniband), I/O counts from the node from the Lustre
-parallel file system, disk I/O counts, cpu and memory utilization
+<p>The reference implementation <i>(hdf5)</i> records I/O counts from the
+network interface (Infiniband), I/O counts from the node from the Lustre
+parallel file system, disk I/O counts, cpu and memory utilization
for each task, and a record of energy use.
<p>The reference implementation stores this data in a HDF5 file for each step
@@ -35,14 +35,14 @@ consolidate all the node-step files in one container for the job.
HDF5 is a well known structured data set that allows different types of
related data to be stored in one file. Its internal structure resembles a
file system with <i>groups</i> being similar to <i>directories</i> and
-<i>data sets</i> being similar to <i>files</i>. There are commodity programs,
+<i>data sets</i> being similar to <i>files</i>. There are commodity programs,
notably <b>HDF5View</b> for viewing and manipulating these files.
<b>sh5util</b> also provides some capability for extracting subsets of date
for import into other analysis tools like spreadsheets.
-<p>This plugin is incompatible with --enable-front-end. It you need to
+<p>This plugin is incompatible with --enable-front-end. It you need to
simulate a large configuration, please use --enable-multiple-slurmd.
-<p>SLURM profile accounting plugins must conform to the SLURM Plugin API with
+<p>SLURM profile accounting plugins must conform to the SLURM Plugin API with
the following specifications:
<p><span class="commandline">const char
plugin_name[]="<i>full text name</i>"</span>
@@ -57,12 +57,12 @@ The minor type can be any suitable name
for the type of profile accounting. We currently use
<ul>
<li><b>none</b>— No profile data is gathered.
-<li><b>io_energy</b>—Gets profile data about energy use and various i/o
+<li><b>hdf5</b>—Gets profile data about energy use and various i/o
sources (local disk, Lustre, network). CPU and memory usage is also gathered.
</ul>
<p>The programmer is urged to study
-<span class="commandline">src/plugins/acct_gather_profile/io_energy.c</span>
-and
+<span class="commandline">src/plugins/acct_gather_profile/acct_gather_profile_hdf5.c</span>
+and
<span class="commandline">src/common/slurm_acct_gather_profile.c</span>
for a sample implementation of a SLURM profile accounting plugin.
<p class="footer"><a href="#top">top</a>
@@ -192,10 +192,10 @@ Put data at the Node Totals level. Typically called when the step ends.
<p style="margin-left:.2in"><b>Arguments</b>: <br>
<span class="commandline">job -- slumd_job_t structure containing information
about the step. </span>
-<br /><span class="commandline">group -- identifies the data stream
+<br /><span class="commandline">group -- identifies the data stream
(source of data). </span>
<br /><span class="commandline">type -- identifies the type of data. </span>
-<br /><span class="commandline">data -- data structure to be put to the file.
+<br /><span class="commandline">data -- data structure to be put to the file.
</span>
<p style="margin-left:.2in"><b>Returns</b>: <br>
<span class="commandline">SLURM_SUCCESS</span> on success, or<br>
@@ -212,10 +212,10 @@ at either job_acct_gather interval or acct_gather_energy interval.
All samples in the same group will eventually be consolidated in one
time series.
<p style="margin-left:.2in"><b>Arguments</b>: <br>
-<br /><span class="commandline">group -- identifies the data stream
+<br /><span class="commandline">group -- identifies the data stream
(source of data). </span>
<br /><span class="commandline">type -- identifies the type of data. </span>
-<br /><span class="commandline">data -- data structure to be put to the file.
+<br /><span class="commandline">data -- data structure to be put to the file.
</span>
<p style="margin-left:.2in"><b>Returns</b>: <br>
<span class="commandline">SLURM_SUCCESS</span> on success, or<br>
@@ -231,18 +231,18 @@ Put data at the Task Totals level. Typically called at task end.
<span class="commandline">job -- slumd_job_t structure containing information
about the step. </span>
<br /><span class="commandline">taskid -- slurm taskid </span>
-<br /><span class="commandline">group -- identifies the data stream
+<br /><span class="commandline">group -- identifies the data stream
(source of data). </span>
<br /><span class="commandline">type -- identifies the type of data. </span>
-<br /><span class="commandline">data -- data structure to be put to the file.
+<br /><span class="commandline">data -- data structure to be put to the file.
</span>
<p style="margin-left:.2in"><b>Returns</b>: <br>
<span class="commandline">SLURM_SUCCESS</span> on success, or<br>
<span class="commandline">SLURM_ERROR</span> on failure.
-<p>Note that the io_energy plugin only uses
+<p>Note that the hdf5 plugin only uses
<i>acct_gather_profile_p_add_sample_data</i>. The job merge program has
-capability for summarizing a time series and inserting grand totals for the
+capability for summarizing a time series and inserting grand totals for the
node. The <i>add_node_data</i> and <i>add_task_data</i> functions were defined
in the intial design and may become depracated.
@@ -265,8 +265,8 @@ It this parameter is not specified, no profiling will occur.
<dt><span class="commandline">ProfileDefaultProfile</span>
<dd>Default setting for --profile command line option for srun, salloc, sbatch.
</dl>
-The default profile value is <b>none</b> which means no profiling will be done
-for jobs. The io_energy plugin also includes;
+The default profile value is <b>none</b> which means no profiling will be done
+for jobs. The hdf5 plugin also includes;
<ul>
<li>
<b>energy</b> sample energy use for the node.
@@ -284,29 +284,29 @@ for the node.
<li>
<b>all</b> all of the above.
</li>
-</ul>
-Use caution when setting the default to values other than none as a file for
+</ul>
+Use caution when setting the default to values other than none as a file for
each job will be created. This option is provided for test systems.
-<p>Most of the sources of profile data are associated with various
+<p>Most of the sources of profile data are associated with various
acct_gather plugins. The acct_gather.conf file has setting for various
sampling mechanisms that can be used to change the frequency at which
samples occur.
<h2>Data Types</h2>
-A plugin-like structure is implemented to generalize HDF5 data operations from
+A plugin-like structure is implemented to generalize HDF5 data operations from
various sources. A <i>C</i> <b>typedef</b> is defined for each datatype. These
declarations are in /common/slurm_acct_gather_profile.h so the datatype are
common to all profile plugins.
<p>
-The operations are defined via structures of function pointers, and they are
+The operations are defined via structures of function pointers, and they are
defined in /plugins/acct_gather_profile/common/profile_hdf5.h and should work
-on any HDF5 implementation, not only io_energy.
+on any HDF5 implementation, not only hdf5.
<p>
-Functions must be implemented to perform various operations for the datatype.
-The api for the plugin includes an argument for the datatype so that the
+Functions must be implemented to perform various operations for the datatype.
+The api for the plugin includes an argument for the datatype so that the
implementation of that api can call the specific operation for that datatype.
-<p>Groups in the HDF5 file containing a dataset will include an attribute for
-the datatype so that the program that merges step files into the job can
+<p>Groups in the HDF5 file containing a dataset will include an attribute for
+the datatype so that the program that merges step files into the job can
discover the type of the group and do the right thing.
<p>
For example, the typedef for the energy sample datatype;
@@ -319,39 +319,39 @@ typedef struct profile_energy {
} profile_energy_t;
</pre>
<p>
-A <i>factory</i> method is implemented for each type to construct a structure
+A <i>factory</i> method is implemented for each type to construct a structure
with functions implementing various operations for the type.
The following structure of functions is required for each type.
<pre>
/*
- * Structure of function pointers of common operations on a
+ * Structure of function pointers of common operations on a
* profile data type. (Some may be stubs, particularly if the data type
* does not represent a time series.
* dataset_size -- size of one dataset (structure size).
- * create_memory_datatype -- creates hdf5 memory datatype
+ * create_memory_datatype -- creates hdf5 memory datatype
* corresponding to the datatype structure.
- * create_file_datatype -- creates hdf5 file datatype
+ * create_file_datatype -- creates hdf5 file datatype
* corresponding to the datatype structure.
- * create_s_memory_datatype -- creates hdf5 memory datatype
+ * create_s_memory_datatype -- creates hdf5 memory datatype
* corresponding to the summary datatype structure.
- * create_s_file_datatype -- creates hdf5 file datatype
+ * create_s_file_datatype -- creates hdf5 file datatype
* corresponding to the summary datatype structure.
- * init_job_series -- allocates a buffer for a complete time
+ * init_job_series -- allocates a buffer for a complete time
* series (in job merge) and initializes each member
- * merge_step_series -- merges all the individual time samples
+ * merge_step_series -- merges all the individual time samples
* into a single data set with one item per sample.
* Data items can be scaled (e.g. subtracting beginning time)
* differenced (to show counts in interval) or other things
* appropriate for the series.
- * series_total -- accumulate or average members in the entire
- * series to be added to the file as totals for the node or
+ * series_total -- accumulate or average members in the entire
+ * series to be added to the file as totals for the node or
* task.
- * extract_series -- format members of a structure for putting
- * to a file data extracted from a time series to be imported into
- * another analysis tool. (e.g. format as comma separated value.)
- * extract_totals -- format members of a structure for putting
- * to a file data extracted from a time series total to be imported
- * into another analysis tool. (e.g. format as comma,separated value.)
+ * extract_series -- format members of a structure for putting
+ * to a file data extracted from a time series to be imported into
+ * another analysis tool. (e.g. format as comma separated value.)
+ * extract_totals -- format members of a structure for putting
+ * to a file data extracted from a time series total to be imported
+ * into another analysis tool. (e.g. format as comma,separated value.)
*/
typedef struct profile_hdf5_ops {
int (*dataset_size) ();
@@ -362,10 +362,10 @@ typedef struct profile_hdf5_ops {
void* (*init_job_series) (int, int);
void (*merge_step_series) (hid_t, void*, void*, void*);
void* (*series_total) (int, void*);
- void (*extract_series) (FILE*, bool, int, int, char*,
- char*, void*);
- void (*extract_totals) (FILE*, bool, int, int, char*,
- char*, void*);
+ void (*extract_series) (FILE*, bool, int, int, char*,
+ char*, void*);
+ void (*extract_totals) (FILE*, bool, int, int, char*,
+ char*, void*);
} profile_hdf5_ops_t;
</pre>
@@ -373,35 +373,35 @@ Note there are two different data types for supporting time series.<br>
1) A primary type is defined for gathering data in the node step file.
It is typically named profile_{series_name}_t.<br>
2) Another type is defined for summarizing series totals.
-It is typically named profile_{series_name}_s_t. It does not have a 'factory'.
+It is typically named profile_{series_name}_s_t. It does not have a 'factory'.
It is only used in the functions of the primary data type and the
-primaries structure has operations to create appropriate hdf5 objects.
+primaries structure has operations to create appropriate hdf5 objects.
-<p>When adding a new type, the <b>profile_factory</b> function has to be
+<p>When adding a new type, the <b>profile_factory</b> function has to be
modified to return an <i>ops</i> for the type.
<p>Interaction between type and hdf5.
<ul>
<li>
-The profile_{type}_t structure is used by callers of the <b>add_*_data</b>
+The profile_{type}_t structure is used by callers of the <b>add_*_data</b>
functions.
</li>
<li>
HDF5 needs a <b>memory</b>_datatype to transform this structure into its
-dataset object in memory. The <i>create_memory_datatype</i> function creates
+dataset object in memory. The <i>create_memory_datatype</i> function creates
the appropriate object.
</li>
<li>
HDF5 needs a <b>file</b>_datatype to transform the dataset into how it will be
written to the HDF5 file (or to transform what it reads from a file into a
-dataset.) The <i>create_file_datatype</i> function creates
+dataset.) The <i>create_file_datatype</i> function creates
the appropriate object.
</li>
</ul>
<h2>Versioning</h2>
-<p>This document describes version 1 of the SLURM Profile Accounting API.
-Future releases of SLURM may revise this API. A profile accounting plugin
-conveys its ability to implement a particular API version using the mechanism
+<p>This document describes version 1 of the SLURM Profile Accounting API.
+Future releases of SLURM may revise this API. A profile accounting plugin
+conveys its ability to implement a particular API version using the mechanism
outlined for SLURM plugins.</p>
<p class="footer"><a href="#top">top</a>
@@ -409,4 +409,3 @@ outlined for SLURM plugins.</p>
<p style="text-align:center;">Last modified 1 April 2013</p>
<!--#include virtual="footer.txt"-->
-