IPv6
Jump to navigation
Jump to search
Google
Description:
The IP protocol used for internet communication is comparable with the zip-code in the postal service industry. An IP-address is an unique identifier, which ensures that you will get the (network) packages you've asked for: e.g. when client A with address_A asks web-server X for webpage Y, then web-server X needs to know client A's address.
With the IPv4 protocol we are out of usable adresses, there are just too many people that want to have a unique address. A new IP-protocol version IPv6 can provide these adresses, and more!
Enablers:
- Increasing internet usage: Severe shortage of IP addresses today
- Increasing mobile usage, pervasive computing: Network resources available worldwide 24/7
- Routing table explosion: IPv6 solves this
- Management Nightmare
- Most IPv4 addresses allocated to United States
- IPv6 has lots of extra features, and is more efficient:
- IPv4 has no support (build in) for new applications (QOS, Mobility etc.)
- IPv4 not scalable, efficient, extensible enough
- IPv4 only has 32-bit addressing
- IPv4 has no security (build in): IPv6 does
- IPv4 has lots of redundancy in protocol
Inhibitors:
- High transistion cost.
- Deployement Issues: Transistioning or direct replacement
- IPv6 is not backwards compatible:
- IPv4 applications cannot work with IPv6. Major patching effort.
- Lack of IPv6 applications in general.
- Transistion phase is needed: Either applications get rewritten, or we translate/tunnel different IP protocols through each other. [IPv4 over IPv6 or IPv4 over IPv6].
Paradigms:
- IPv4 era: Limited amount of online users thus a limited amount of IP-addresses, mostly allocated to US.
- IPv6 era: Enormous boom in online users, and new (mobile) networks demand for more and more IP-addresses. Backwards compatability should be solved by workarounds like tunneling or translation.
Experts:
Timing:
When the demand for new IP-addresses is high enough to force the industry to invest in IPv6 infrastructure.