HP XC System Software Administration Guide Version 2.1

Environment Variable Description
LSF_HPC_NCPU_INCREMENT=increment
This is the upper limit for the number of CPUs
that changed since the last checking cycle. The
default value is 0 (zero).
LSF_HPC_NCPU_INCR_CYCLES=incr_cy-
cles
The incr_cycles parameter is the minimum
number of consecutive cycles where the
number of CPUs changed does not exceed
LSF_HPC_NCPU_INCREMENT. The default
value is 1. LSF-HPC checks total usable CPUs
every two minutes.
LSF_HPC_NCPU_COND= or | and
This setting defines how two thresholds are
combined. The only valid values are and and or.
The default is or.
12.12 Extending the LSF-HPC Cluster
The following sections describe ho w to extend the LSF-HPC.
Be sure to see the Useful links portion of the HP XC Clusters Web site:
http://www.hp.com/techservers/clusters/xc_clusters.html
Be sure to select the appropriate link for yo ur release of the HP XC System Software to f ind
HowTos and o ther docum entation on recent developments on this and other topics.
12.13 Configuring an External Virtual Hostname for LSF-HPC
on HP XC Systems
An external virtual hostnam e for LSF-H PC is necessary when LSF-HPC needs to be accessed
from the external network. This access could be from non-XC LSF client hosts, or from other
LSF clusters when the HP XC cluster is configured in a MultiCluster configuration. See the
LSF docum entation for more details.
Perform the following steps to configure an external virtual h ostn ame:
1. If LSF-HPC failover is enabled, ensure t hat each node with the resource_management role
has an external network connection. Run the following command to confirm this:
# shownode roles --role resource_management external
resource_management: n[127-128]
external: n[125-128]
If necessary, fixing this is beyond the scope of these procedures.
2. Obtain a virtual IP and correspo nding hostname, and ensure that they are not already in use:
# ping xclsf
resource_management: n[127-128]
external: n[125-128]
3. Ensure th at no job s are running. Checking for running j obs and draining the nodes (if
necessary) are beyond the scope of these procedures.
4. Shut down LSF:
# controllsf stop
5. Change the virtual hostname to the new one and confirm:
# controllsf set virtual hostname xclsf
# controllsf show
LSF is currently shut down, and assigned to node .
Failover is disabled.
Head node is preferred.
The primary LSF host node is n128.
12-14 LSF-HPC Administration