I am curious on how others handle making network maps? I remember reading somewhere that multiple maps should be drawn, which reflects different layers of the OSI model. Example: overall network map (high level view), Layer 3 and Layer 2 maps. In addition, when dealing with Layer 2 and STP – is it beneficial to map the physical network, and then map the ports that are in blocking status on a different map?
Since I am very new to the industry – how do most network veterans handle their network maps? What should I focus on? What should I pay attention too and not get slowed down?
I've heard the same argument about multiple maps for the varying OSI layers. While it's a good idea, it often falls apart because the people advocating it in theory aren't usually the ones actually maintaining it. Or its created as part of a major refresh or Greenfield deployment and then staff aren't assigned to own the ongoing maintenance of it.
For those reason, I say start small - a map that at least has all of your devices and the relevant connectivity between them is a good start. Add in some VLAN and addressing as necessary to add value to the staff who use the map.
Most importantly no matter what scheme you decides best suits you is to have a process to keep the information current. Without that, any theoretical advantage of one representation vs. another quickly becomes a moot point.