
5.4
Launch an HPC Job Using mrun
Prerequisites
Use the
urika-state
command to ensure that both the Mesos and Marathon services are running and to check
if the system is operating in the service mode that supports using
mrun
. For more information, see the
urika-
state
man page and refer to
About this task
The
mrun
command can be used to launch HPC jobs on the Urika-GX system.
The
mrun
command cannot be executed within a tenant VM
Procedure
Launch the HPC job using the
mrun
command, specifying the number of nodes to allocate.
In the following example,
my_hpc_app.exe
is used as an example for the name of the application to run.
$
mrun -n 32 -N 8 my_hpc_app.exe arg1 arg2 arg3
Refer to the
mrun
man page for more information, further examples, environment variables, configuration files
and command-line option descriptions of the
mrun
command.
5.5
Manage Long Running Services Using Marathon
Marathon is used by the Cray-developed command named
mrun
to allocate node resources from Mesos and
launch application instances as needed. In addition, Cray-developed scripts for starting a cluster of YARN Node
Managers are also launched through Marathon.
Before using Marathon, ensure that the system is running in the service mode that allows use of this service.
Execute the
urika-state
or
urika-service-mode
commands to check the service mode. For more
information, refer to the
urika-state
or
urika-service-mode
on page 177.
CAUTION: Unless it is required to shut down YARN nodes, analytic applications that use the Cray-
developed scripts for flexing a cluster should not be deleted through the Marathon UI, as doing will lead to
loss of YARN nodes.
On the Urika-GX system, there are always three Mesos Masters and three Marathon instances running, while one
of them is the active leader. Requests received by the login node are proxied to the currently active leader. If a
leader runs into issues, one of the backup leaders take over and the requests are proxied to the current leader.
Access the Marathon web UI by selecting Marathon on the Urika-GX Applications Interface, located at:
http://
hostname
-login1
. Though this is the recommended method of accessing Marathon, it can also be
Resource Management
S3016
133