Monitoring a single device with dozens of interfaces?

[Capacitor]

How does Zabbix monitor devices with dozens (or even hundreds) of interfaces, and how is this information represented on a graph?

Is it one "service check" per interface? ie: does it have to connect to the device once each time per interface?
Is it one SNMP query per interface, or can this be consolidated into one connection/query for all interfaces?

How is this displayed in graphs? is it one graph per interface, is it one massive graph with hundreds of little lines?

Is there the ability to have a 10,000 ft view of the whole device - which is what we network admistrators (and others prefer) ?

Several examples of devices to monitor are like the Cisco AS5300, AS5400, and AS5800.
The AS5800 can have close to 800 connections, which all need to be monitored.

MRTG comes closest to doing this, but isn't designed really to do much else.

What's the best recommendation for this?

[mucknet]

See the following thread:

http://www.zabbix.com/forum/showthread.php?t=5726

Unfortunately its not a fix for your problem, just letting you know that others want the same thing

[Capacitor]

It's not just indexed tables

It's not just indexed tables, because often times the devices don't have those.

It's also about how this information is presented. Graphing.

MRTG can do a very good job for a single device - or at least it can be made to, to create an "index" page, and then create (manually) a master index page of those index pages.

It's not enough just to consider how the information is to be collected (we could use Cacti for that), it's how we're going to display it.

Having a graph with 800 lines on it doesn't help anyone.