<html>
<head>
<title>
February 19, 2000 JSD 1
</title>
<style type="text/css">
<!--
body { margin-left:4%; }
H1, H2, H3, H4, H5 {
color: maroon; padding: 4pt; margin-left: -4%;
border: solid; border-width: thin; width: 100%;
background: rgb(204,204,255)
}
-->
</style>
</head>
<body bgcolor="#FFFFFF" text="#000000">
<h3>
NAME
</h3>
<b>jsd</b>
- simple command scheduling daemon for remote execution
<h3>
SYNOPSIS
</h3>
<b>jsd</b>
[<b>-</b><b></b><b>iqv</b>]
[<b>-</b><b></b><b>f</b><i></i><i> fanout</i>]
[<b>-</b><b></b><b>g</b><i></i><i> rungroup1,...,rungroupN</i>]
[<b>-</b><b></b><b>l</b><i></i><i> username</i>]
[<b>-</b><b></b><b>p</b><i></i><i> port</i>]
[<b>-</b><b></b><b>o</b><i></i><i> port</i>]
[<b>-</b><b></b><b>w</b><i></i><i> node1,...,nodeN</i>]
[<b>-</b><b></b><b>x</b><i></i><i> node1,...,nodeN</i>]
[<i></i><i>command</i><i> ...</i>]
<h3>
DESCRIPTION
</h3>
The
<b>jsd</b>
daemon is used to schedule remote execution of jobs across a
heterogenous cluster of machines. Upon starting the daemon, it will
attempt to run a benchmark on all machines under it's control.
Once the daemon is active it listens for connections from a
<em></em><em>jsh</em>
process, and gives each connecting process the name of a remote host
to execute a command on. Faster hosts are preferred by the
<b>jsd</b>
daemon, and the fastest of all remaining hosts will allways be given
out first. Once a remote
<em></em><em>jsh</em>
process has completed it's use of the remote host, it will again
contact the
<b>jsd</b>
daemon, and return the node to service. The daemon will never give
the same node out to more than one process at a time.
<p>
<b>jsd</b>
logs errors and critical conditions to the syslog with the facility
<code></code><code>LOG_DAEMON</code>.
<p>
The following options are available:
<dl compact>
<p><dt><b>-</b><b></b><b>i</b><dd>
The
<b>-</b><b></b><b>i</b>
option will list information about the current cluster, and command groupings.
It will print out the current value of the fanout, and how many groups of
machines there are within the cluster. It will also show you which command
you are about to run, and your username if specified with the
<b>-</b><b></b><b>l</b>
option.
<p><dt><b>-</b><b></b><b>q</b><dd>
The
<b>-</b><b></b><b>q</b>
option does not issue any commands, but displays information about the
cluster, and the fanout groupings.
<p><dt><b>-</b><b></b><b>v</b><dd>
Prints the version of ClusterIt to the stdout, and exits.
<p><dt><b>-</b><b></b><b>f</b><dd>
If the
<b>-</b><b></b><b>f</b>
option is specified, followed by a number, it sets the fanout size of the
cluster. The fanout size is the number of nodes a command will run on in
parallel at one time. Thus a 80 node cluster, with a fanout size of 64,
would run 64 nodes in parallel, then, when all have finished, it would
execute the command on the last 16 nodes. The fanout size defaults to 64.
This option overrides the
<code></code><code>FANOUT</code>
environment variable.
<p><dt><b>-</b><b></b><b>g</b><dd>
If the
<b>-</b><b></b><b>g</b>
option is specified, followed by a comma separated list of group names, the
command will only be run on that group of nodes. A node may be a part of
more than one group if desired, however running without the
<b>-</b><b></b><b>g</b>
option will run the command on the same node as many times as it appears in the
file specified by the
<code></code><code>CLUSTER</code>
environment variable. This option is silently ignored if used with the
<b>-</b><b></b><b>w</b>
option.
<p><dt><b>-</b><b></b><b>l</b><dd>
If the
<b>-</b><b></b><b>l</b>
option is specified, followed by a username, the commands will be run under
that userid on the remote machines. Consideration must be taken for proper
authentication, for this to work.
<p><dt><b>-</b><b></b><b>p</b><dd>
The
<b>-</b><b></b><b>p</b>
option can be used to set the port number which
<b>jsd</b>
will listen to
replies from
<em></em><em>jsh</em>
processes indicating completion of processing on remote nodes. This
option overrides the
<code></code><code>JSD_IPORT</code>
environment variable. The default listen port is 2001.
<p><dt><b>-</b><b></b><b>o</b><dd>
The
<b>-</b><b></b><b>o</b>
option can be used to set the port number which
<b>jsd</b>
will listen to
requests from
<em></em><em>jsh</em>
processes for remote nodes. This option overrides the
<code></code><code>JSD_OPORT</code>
environment variable. The default listen port is 2002.
<p><dt><b>-</b><b></b><b>w</b><dd>
If the
<b>-</b><b></b><b>w</b>
option is specified, followed by a comma delimited list of machine names,
the
<i></i><i>command</i>
will be run on each node in the list. Without this option,
<b>jsd</b>
runs on the nodes listed in the file pointed to by the
<code></code><code>CLUSTER</code>
environment variable.
<p><dt><b>-</b><b></b><b>x</b><dd>
The
<b>-</b><b></b><b>x</b>
option can be used to exclude specific nodes from the cluster. The format
is the same as the
<b>-</b><b></b><b>w</b>
option, a comma delimited list of machine names. This option is silently
ignored if used with the
<b>-</b><b></b><b>w</b>
option.
</dl>
<h3>
ENVIRONMENT
</h3>
<b>jsd</b>
utilizes the following environment variables:
<dl compact>
<p><dt><code></code><code>CLUSTER</code><dd>
Contains a filename, which is a newline separated list of nodes
in the cluster.
<p><dt><code></code><code>RCMD_CMD</code><dd>
Command to use to connect to remote machines. The command chosen must
be able to connect with no password to the remote host. Defaults to
<b></b><b>rsh</b>.
This is only used by
<b>jsd</b>
when issuing the benchmark at startup, and has no
effect on how
<em></em><em>jsh</em>
processes will connect to the remote nodes.
<p><dt><code></code><code>RCMD_CMD_ARGS</code><dd>
Arguments to pass to the remote shell command. Defaults to none. As with
<code></code><code>RCMD_CMD</code>
this is only used when issuing the benchmark at startup.
<p><dt><code></code><code>JSD_BENCH_CMD</code><dd>
This is the name of a command, which when run on the remote nodes of
the cluster will give some indication of the relative speed of the
machines. It does not need to be precise. If no benchmark command is
specified, all machines are considered to be the same speed, and will
not be wieghted. The bench command expects a single floating point
number as the only output to the remote command. Larger numbers are
to indicate faster machines. See the
<em></em><em>EXAMPLES</em>
section for more details.
<p><dt><code></code><code>FANOUT</code><dd>
When set, limits the maximum number of concurrent commands sent at once.
This can be used to keep from overloading a small host when sending out
commands in parallel. Defaults to 64. This environment setting can be
overridden by the
<b>-</b><b></b><b>f</b>
option.
<p><dt><code></code><code>JSD_OPORT</code><dd>
When set, overrides the default port which
<b>jsd</b>
listens on for
<em></em><em>jsh</em>
processes requesting nodes. This environment setting can be overriden
by the
<b>-</b><b></b><b>o</b>
option.
<p><dt><code></code><code>JSD_IPORT</code><dd>
When set, overrides the default port which
<b>jsd</b>
listens on for
<em></em><em>jsh</em>
processes reporting back that it has completed processing on a node.
This environment setting can be overriden
by the
<b>-</b><b></b><b>p</b>
option.
</dl>
<h3>
EXAMPLES
</h3>
In order to generate a benchmark figure useful to
<b>jsd</b>
you will most likely need to process the ouput of a benchmark command
in some way. There are two simple ways of doing this:
<p>
The first way is to build a script, which exists on each machine, and
run that as your benchmark command. The below script shows how to
process the output of the
<a href="whetstone.html">whetstone(1)</a>
command into a format useable by
<b>jsd</b>
:
<dl compact><dt><dd>
<code>
<pre>
#!/bin/sh
./whetstone 100000 | grep MIPS | sed -e 's|.*: \(.*\) MIPS|\1|'
</pre>
</code>
</dl>
<p>
The other way, would be to run a more involved benchmark on each
machine, (or assign numbers arbitrarily) and place them in a file on
the remote machines. A benchmark command could then simply be
<b></b><b>cat</b><b> /etc/benchmark</b>
or some other similar setup.
<h3>
SEE ALSO
</h3>
<a href="dsh.html">dsh(1)</a>,
<a href="jsh.html">jsh(1)</a>,
<a href="../html8/syslogd.html">syslogd(8)</a>.
<h3>
HISTORY
</h3>
The
<b>jsd</b>
command first appeared in clusterit 2.0. It should be pronounced
"jez-dee,
and stands for
`Job Scheduling Daemon.'
<h3>
AUTHOR
</h3>
<b>Jsd</b>
was written by Tim Rightnour.
</font></body>
</html>
distrib
>
Mandriva
>
2010.0
>
i586
>
media
>
contrib-release
>
by-pkgid
>
24e64c3909f783dd991f17e423e8acf5
>
files
>
26
