This is a question more for the ISP side. I'm trying to develop some reporting at the node level for a HFC ISP, and I want to make sure I'm not overlooking any useful metrics for diagnosing node health. Right now the main metrics that I track are committed bandwidth to the node, oversubscription rate, maximum utilization for last 7 days, and average utilization for the last 7 days.
Are there any other related metrics that would be useful for a networking person to see which nodes are 'healthy' (or unbalanced, or overloaded, etc.)? Assume I can deploy probes on about anything (they don't use Nagios, but they use Cacti) to monitor it.
Thank you for any insights you can offer!
No comments:
Post a Comment