Troubleshooting guide

and on host2 type:

gm_allsize --both-ways --bandwidth \

--remote-host=host1 --size=15 –geometric

where the length of the messages sent is 2**(size - 8) bytes. This test has GM

streaming packets in both directions (both nodes are always sending) and it causes

GM to report the sum of the send and receive bandwidths.

The output from this command will consist of two columns of data: the first

column lists the message size (in bytes) and the second column lists the

bandwidth (in MB/s).

4. Run a sample benchmark (e.g., HPL) (1 node run) on each of the nodes in the

cluster to ensure that all nodes report consistent performance. If not, there could

be an issue with a particular CPU on one of the hosts.

5. Run a sample benchmark (e.g., HPL) on equally-sized subsets of nodes. Make

sure that performance is consistent across all subsets of nodes. If you see a

particular subset that is slower, then you need to perform a divide-and-conquer

approach to isolate the slower node.