CPU usage high

[wuhaohust]

I configured this agent on HUB (AIX) to check the non-response agent from two RTEMS, when the agent run the perl script to collect the agent status. I found each instance used around 7~9% CPU ,two instance took about 18% . And it took more than 15 mins to completed the scanning . Agent on each RTEMS is less than 1000. We totally have more than 20 RTEMSs,so wonder if I configured the agent for all the RTEMSs, it may crash the HUB. Any idea ?Thanks in advance!