(“In Case You Missed It Monday” is my chance to showcase something that I wrote and published in another venue, but is still relevant. This week’s post originally appeared on Orange Matter)
“Once upon a time, there was a little network…”
No there wasn’t. Your network was never little, or if it was, it didn’t stay that way for long. Corporate environments grow, often in unexpected ways. This can cause IT professionals to weep bitter tears of frustration in the cold dark midnight of their change control windows. But there is a more basic cause for this feeling. It’s not just that networks have grown larger, or even more complex. The fundamental nature of what a “network” is has changed, and the tools intended to monitor and manage it have, in a large part, not.
A decade or so ago, “the network” was a collection of routers and switches. Period. Ok, throw in an access point or two, but honestly, that was about it. Monitoring and managing the network was a matter of ensuring that those two basic device types were up, their sub-elements were doing their job, and that the packets flowed freely.
That’s not to say that networks were simple. Even then they were a complex beast, sonnet and token ring and competing L2/L3 protocols. But much of what made it complex back then has been sorted and simplified today, and we’ve moved on to a different type of complexity.
Today, your network still appears to operate under the old paradigm. There are bits and packets and frames, and data moves through devices and cables like a well-plumbed water park. But there are specialty devices now: load balancers and firewalls and data center fabric switches. These devices may appear to operate similarly to their router and switch forebears, but in reality are nothing like them.
What sets these devices apart is that they are relatively few in number and that they serve in a capacity that is both highly unique and highly critical. But what can be maddening to the network engineer is that these devices DO resemble a router or a switch—at least enough that monitoring and management software have traditionally chosen to leave their statistics as-is and call it good enough, never addressing those specialized behaviors.
In this series, my goal is to lay out, at a high level, the basic nature of these devices and their essential functions; and to point out how that basic nature and essential functions differ from the statistics that monitoring and management tools typically gather.