| Ron Bennett |
09-18-2005 08:55 PM |
The root server data is cached throughout much of the Internet - ie. most ISPs, servers, etc cache that data.
The zones, such as .COM, .NET .CO.UK, .DE, etc are another story completely - but again, zones are widely distributed / cached.
In a nutshell, while many folks focus on root servers, etc ... that's among the stronger, more resilient parts of the internet despite how centralized that part is.
Routing, in particular NAPs, is where the internet is more vulnerable ...
Zap a few NAPs (network access points where large amounts of traffic is exchanged with other networks) and much of the internet will grind to a halt; peering arrangements outside NAPs only offers minimal protection in situations where numerous NAPs are affected due to the large volume of traffic involved.
Continuing on that thought - hacking, etc likely won't do diddly overall ... for a NON-government entity to knock out the internet would likely require physical access / destruction to facilities/networks ...
In regards to the government ... they can shut off large portions of the internet (most, if not all, NAPs are already tapped by governments) any time if the government was determined too; they likely never would because much of the government itself greatly relies on the same internet network itself for critical functions - ie. military, nuclear plants, collecting taxes, banking (some parts, such as Fedwire, are totally on private networks, but ATMs, on-line banking, etc are not) ...
Contrary to belief of some, Internet II is very limited in scope; much infrastructure relies on the same internet folks use ... anyone doubting that need look no further than the Slammer worm that led, in part, to the blackouts in 2003.
Ron
|