Actually, it is even worse than that, because active sites like www.facebook.com do not have only one network number.

Large net properties could have stands and racks of customer-facing servers in processes centres all around the globe, giving them a wide selection of network number ranges on a wide selection of distinct networks.

Active websites commonly use DNS to direct you distribute load to a particular server predicated on load levels, maintenance schedules, your present place, etc, as a way to enhance rate and prevent bottlenecks.

To put it differently, DNS is very significant, to the point the web would be unusable without it.

As an example, to determine where nakedsecurity.sophos.com resides, your own firm’s (or ISP’s) DNS server takes a top down strategy:

Request the so called root servers, “Who looks after the .COM domain name information?”
Question the .COM portion of the hierarchy, “Who’s formally responsible for DNS for SOPHOS?”
Each DNS response has TTL, that says to recall the response, usually from 1 minute to 1 hour, after, or a Time To Live amount.

That significantly reduces the amount of times a complete, top down hierarchical query is required, while ensuring the system can recover automatically from out-of-date or wrong responses.

As it is possible to picture, the root servers are the key to the whole DNS service, because all as yet-unknown responses have to be requested by beginning at the top.

So there are 13 root servers named A to M, managed by 12 distinct organisations, on 6 different continents.

The truth is, each “server” really is made up of server farm of several real servers in multiple places, for dependability.