FIT1047 Chap.10 IP Addressing, Routing, TCP and UDP
IP Addressing, Routing, TCP and UDP
IP addresses are interpreted with prefix lengths that define network membership and route blocks. Hosts compare destinations with local prefixes, then select a direct neighbour or router next hop. Routers use longest-prefix matching to choose the most specific route and rebuild link context for each hop. NAT can translate address/transport identifiers under state but is not equivalent to security policy.
Transport ports demultiplex traffic to processes: TCP supplies a connection-oriented ordered byte stream with sequence, acknowledgement, retransmission, flow and congestion behaviour, while UDP supplies discrete datagrams without TCP's ordering/retransmission guarantees. Application requirements determine which semantics fit.
This chapter combines prefix, forwarding and transport evidence so a matching route, host arrival and application response are never treated as the same outcome. Examples are original standard canon within published networking scope and do not predict the in-class practical or test.
The diagnostic chain should preserve distinct decisions: address interpretation, local-prefix comparison, route selection, next-hop delivery, transport demultiplexing and application handling. Evidence at one stage narrows but does not settle later stages. Longest-prefix matching is a routing rule, whereas link resolution and NAT state are separate mechanisms.
This is original standard canon within the published A3 purpose; no in-class task layout, address plan, duration, marking scheme, tool set or permitted-resource rule is reconstructed.
What this chapter covers
- 01
IP addresses plus prefix lengths
- 02
Network blocks and membership
- 03
Host local-versus-router decision
- 04
Longest-prefix route lookup
- 05
Next hops and per-hop frames
- 06
NAT translation state
- 07
Ports and socket demultiplexing
- 08
TCP stream versus UDP datagram semantics
- 09
Evidence levels from route to application
- 10
Transport endpoint binding and demultiplexing
- 11
Keeping NAT state separate from policy
AskSia-authored practice weighting (not an official mark scheme): Fresh route-and-service decision
- routeList all matching prefixes and select the department route as the longest match.
- forwardUse its next hop/outgoing interface and build a local frame to that next hop.
- packetRetain the remote IP destination while link addresses change at each router.
- transportAt the host, use UDP protocol and destination port for process delivery.
- classifyConclude network delivery can succeed while application delivery fails because no process is bound.
Key terms
- Prefix length
- The count of leading address bits fixed for a network block.
- Longest-prefix match
- Selection of the most specific matching route for a destination.
- Next hop
- The immediate router/interface destination selected for the current forwarding step.
- NAT
- Stateful translation of network and sometimes transport identifiers across a boundary.
- Port
- A transport-layer identifier used with addresses/protocol state to deliver to a process.
- Byte stream
- TCP's ordered sequence abstraction without preserving application write boundaries.
- Demultiplexing
- The process of using transport and network identifiers to deliver received data to the correct endpoint context or application. Successful host delivery does not guarantee that a process is bound to the destination port or prepared to handle the message.
- Route-selection evidence
- Evidence that a destination matched a particular prefix and next-hop decision. It does not by itself establish link delivery, remote-host acceptance, transport availability or application success, which require later observations.
IP Addressing, Routing, TCP and UDP FAQ
Can I decide whether two hosts are local from addresses alone?
You need the relevant prefix lengths and topology. Applying the same mask can establish prefix membership, not guaranteed connectivity or policy.
Why does the longest route win instead of the first?
The longer matching prefix describes a smaller, more specific destination block. Default is least specific and applies only without a better match.
Is NAT a firewall?
No. NAT translates identifiers and tracks mappings. A firewall separately enforces traffic policy, though one device may implement both.
Does TCP preserve messages?
No. It supplies ordered bytes. The application must frame its own messages because writes can be split or combined across segments.
When is UDP appropriate?
When datagram boundaries, timeliness, low overhead or application-controlled recovery fit requirements. The application still owns congestion-aware behaviour and any needed reliability.
Why can a correct route coexist with a failed application request?
Routing selects a next hop for the destination prefix. Later stages can still fail at neighbour delivery, host policy, transport binding, protocol exchange or application logic. The diagnosis should preserve the route result while testing the next unresolved event.
What A3 details remain deliberately unknown in this chapter?
The exact applied-networking activity, in-class test structure, address plan, tools, duration, reflection prompt, rubric and permitted materials. Only the published A3 purposes are used; operational details must be read from current Moodle.
Assessment move
For prefix practice, mark fixed bits, compute aligned block boundaries and verify membership with a mask. Build route tables with prefix/length/next hop/interface and circle the longest matching row before any link resolution. Trace two routers with three different link frames around the continuing IP packet. Add NAT only with before/after tuples and reverse mapping state.
For transport, use flow tuples and separate TCP setup, byte acknowledgement and application parsing. For UDP, record message identifiers, loss/duplicate possibilities and application retry semantics. Build a hypothesis ladder from route match to next-hop delivery, host arrival, transport endpoint and application response. Use only authorised probes and leave silent-path causes open until evidence discriminates them.
Treat textbook prefix, routing and transport material as conceptual support, not reusable answer text. Build fresh route tables, tuples and packet paths, and never copy source diagrams or solved examples. Use authorised probes only and follow current Moodle for the actual A3 analysis, reflection and test conditions.
Working through IP Addressing, Routing, TCP and UDP in FIT1047? Sia is AskSia’s AI Computer Science tutor — ask any FIT1047 IP Addressing, Routing, TCP and UDP question and get a clear, step-by-step explanation grounded in how FIT1047 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
