HP XC System Software Installation Guide Version 2.1

The m anagement database names of the services provided by this role are:
•SLURMcompute(slurm_compute)
•SuperMon(mond)
• Nagios remote plugin execution agent (nrpe)
To enable monitoring of the nodes, nodes with the compute role runs the Nagios Remote
Plugin Execution agent and the SuperMo n Monitoring service.
Any node with this role may be called upon to execute jobs sch e dul ed by the users. Nodes with
this role are also o ft en called compute nodes.
I.4.2 Login Role
Nodes with the login accept login sessions of users. A user can submit jobs from the
command line on a node with a login role. The jobs then are distributed among compute nodes
to process the job. Nodes with this role should have an external network connection configured.
The m anagement database nam e for th e service supplied by t his role i s lvs.
The login role supplies a node with the LVS Director service, wh ich handles the placement of
user login sessions on login nod es wh en a u ser logs in to the cluster alias.
I.4.3 I/O Role
Nodes with the I/O role provide access to storage and file systems m ounted locally on the
node. This role can be located on any n ode that provides local file system access to all nodes
using NFS. Assign this role to any node that is exportin g S AN storage.
The managem ent database name for the NFS Server service supplied by this role is
nfs_server.
Nodes with t his role normally reside in the u t ili ty cabinet of the cluster and have the most
direct access to storage.
Yo u can assign other roles to a node with this role. However, you m ust be careful not to
overload the node so it can provide adequate NFS service.
I.4.4 Management Hub Role
A node w ith the management hub role is an aggregation point fo r management activities.
The m anagement database names of the services provided by this role are:
• Myrinet switch monitor (gmmon)
• Apache Web server (httpd)
• Nagios aggregator (nagios)
• Nagios service ( nsca)
• SuperMon aggregator (supermond)
• syslog-ng aggregator (syslogng_forward)
• Quadrics switch monitor (swmlogger)
These services are used to support scaling out of the cluster. Nod es with this role provide lo cal
storage for aggregation of system logs and performance info rm ation for the supermond and
syslogng_forward services.
Yo u can assign this role to several nodes of the cluster. It is assigned to the head no de by default.
I-4 Using the Configuration Menu