beautypg.com

HP XC System 3.x Software User Manual

Page 69

background image

NOTE:

The --nodelist=nodelist option is particularly useful for determining

problematic nodes.

If you use this option and the --nnodes=n option, the --nnodes=n option is ignored.

test

Indicates the test to perform. The following tests are available:
cpu

Tests CPU core performance using the Linpack benchmark.

cpu_usage

Tests CPU core usage. All CPU cores should be idle during
the test. This test reports a node if it is using more than
10% (by default) of its CPU cores.

The head node is excluded from this test.

memory

Uses the streams benchmark to test memory performance.

memory_usage

Tests memory usage. This test reports a node that uses
more than 25 percent (by default) of its memory.

network_stress

Tests network performance. Check network performance
under stress using the Pallas benchmark's Alltoall,
Allgather, and Allreduce tests. These tests should be
performed on a large number of nodes for the most
accurate results.

The default value for the number of nodes is 4, which is
the minimum value that should be used.

The --all_group option allows you to select the node
grouping size.

network_bidirectional

Tests network performance between pairs of nodes using
the Pallas benchmark's Exchange test.

network_unidirectional

Tests network performance between pairs of nodes using
the HP MPI ping_pong_ring test.

You can list the available tests with the ovp -l command:

$ ovp -l

Test list for perf_health:

cpu_usage

memory_usage

cpu

memory

network_stress

network_bidirectional

network_unidirectional

By default, the ovp command reports if the nodes passed or failed the given test. Use the ovp --verbose
option to display additional information.

The results of the test are written to a file in your home directory. The file name has the form
ovp_node_date[rx].log

where node is the node from which the command was launched and date

is a date stamp in the form mmddyy. Subsequent test runs of the test are identified with an r (for run) and
a number; this prevents a log file from overwriting a previous one. The --chdir= option enables you to
designate a directory different from your home directory as the execution directory.

The following is an example of the CPU usage test:

7.6 Running Performance Health Tests

69