I've spent the last two weeks (or most of it) going over network management and equipment record systems. The problem I found is that something simple almost always ends up being un-usefull. Something that is very usefull almost always ends up being overly-complex. There was no happy medium. It is somewhat understandable given the complexity of the task at hand.

I ended up using Nagios but my needs are somewhat different from yours. I was looking for a solution that could do active, passive, and SNMP monitoring of networks. It also had to be able to give me dependency and relationships up and down for incident ticketing. In other words, if it detects a problem with a server, it creates a ticket for just that server/client. If it detects a problem with a switch, it creates a ticket for the switch and all servers/clients attached to the switch.

Most of the stuff I know for mapping is equipment records in network monitoring programs. Don't really know of anything to parse an excel spreadsheet and make them.