Computer Network & Network Security System β Network Layer, NEC licence examination syllabus (Nepal Engineering Council).
4.3 billion IPv4 addresses ran out. IPv6 offers roughly 340 undecillion β deliberately, absurdly overkill.
Address size128 bits (vs IPv4's 32 bits), written in hexadecimal groups (e.g. 2001:0db8::1).
Simplified headerFewer fields than IPv4's header, designed for faster router processing.
Extension headersOptional headers (for security, routing options, fragmentation, etc.) chained after the main header only when actually needed β instead of IPv4's approach of always reserving space for options that are usually unused.
Dual stackA device runs both IPv4 and IPv6 simultaneously, using whichever the destination supports.
TunnelingIPv6 packets get wrapped inside IPv4 packets to cross IPv4-only network segments.
Header translationActively rewrites headers between the two formats at the boundary between an IPv4 and IPv6 network.
IPv6 multicasting is built into the protocol design from the ground up (IPv4 mostly bolted multicast on later) β a single IPv6 address prefix is dedicated entirely to multicast use.
"340 undecillion" conveys nothing useful. Two comparisons make it concrete.
No header checksumRemoved entirely. The link layer already checks frames and the transport layer checks end to end, so recomputing it at every router was pure cost β a router must recompute it because it decrements the hop limit.
No router fragmentationOnly the sender may fragment. A router that meets a too-large packet discards it and reports back, so routers do no per-packet reassembly bookkeeping.
Fixed 40-byte headerOptions moved into optional extension headers, so the common case is a fixed size a router can parse without examining lengths.
IPv6 is not backward compatible β an IPv6-only host cannot talk to an IPv4-only host, because the address does not fit in the field. So the two run side by side, with hosts and routers supporting both, and neither side can switch off until essentially everyone has moved.
The other reason is that NAT worked too well. Private addressing let one public address front thousands of machines, which relieved the pressure that would otherwise have forced migration. It is a workaround that removed the urgency of the fix β at the cost of end-to-end addressability, which is why protocols like AH break through NAT and why peer-to-peer connections need workarounds.Create a free account to tick topics off, take notes as you read, watch the video lessons and get a day-by-day study plan built around your exam date.
Loadingβ¦