Which type of Cisco address translation allows thousands of internal hosts to share a single public IP address by using different TCP and UDP port numbers for each session?

Introduction / Context:
Public IPv4 addresses are scarce, so most organisations and almost all home networks cannot assign a unique public IP to every internal device. Instead they rely on a form of NAT that lets many internal hosts share a single public address by multiplexing connections through different port numbers. This approach is called Port Address Translation, or PAT, and is implemented with the overload keyword in Cisco configurations.

Given Data / Assumptions:

There is only one or a very small number of public IP addresses available.

You need to allow potentially thousands of internal hosts to access the Internet.

The translation method must distinguish sessions by port numbers, not only by IP addresses.

Cisco IOS offers static NAT, dynamic NAT and PAT or NAT Overload as options.

Concept / Approach:
Static NAT creates a permanent one to one mapping between an inside local and an inside global address, which cannot scale to thousands of hosts sharing one address. Dynamic NAT uses a pool of inside global addresses and maps inside locals to them as needed, but each mapping still consumes one public address. PAT or NAT Overload, by contrast, maps multiple inside local addresses to a single inside global address by assigning unique source port numbers for each active session. This allows thousands of simultaneous translations using only one or a few public IPs.

Step-by-Step Solution:
Recognise the requirement: thousands of hosts must share one address, which rules out pure static NAT and simple dynamic NAT without port reuse. Recall that PAT is configured with commands like ip nat inside source list 1 interface serial0 overload, where overload is the key to sharing a single address among many hosts. Under PAT, the router keeps track of each internal host and session by using a combination of the inside global address and a unique source port, allowing many internal addresses to appear to come from the same public IP. This behaviour matches the description given in the question exactly. Therefore, Port Address Translation using NAT Overload is the correct translation type.

Verification / Alternative check:
You can verify PAT behaviour with show ip nat translations, which will show multiple entries sharing the same inside global address but using different port numbers, such as tcp 203.0.113.5:30001 mapping to 192.168.1.10:1025 and tcp 203.0.113.5:30002 mapping to 192.168.1.11:1025. This demonstrates how a single public IP can represent many internal hosts simultaneously, which is characteristic of PAT.

Why Other Options Are Wrong:
Option a, static one to one NAT, assigns a dedicated global address for each internal host and cannot scale to thousands of hosts with only one public IP.

Common Pitfalls:
Option b, dynamic many to many NAT using a pool, still requires a public address for each concurrent internal host and therefore does not allow thousands of hosts to share a single address.
Option d, policy based NAT, refers to NAT decisions that depend on policies such as source or destination, but it does not inherently address the many to one scaling described here.
Option e, outside source static NAT, maps external addresses into the inside network for special use cases and does not match the scenario of internal hosts going out through one public IP.

Common Pitfalls:
A common misunderstanding is equating any NAT configuration with PAT. Only configurations that use overload or explicit port mapping behave as PAT, allowing large numbers of hosts to share a small number of addresses. Knowing which NAT type you are using is crucial when diagnosing issues with simultaneous connections, port exhaustion or address shortages.

Final Answer:
The translation type that allows thousands of hosts to share a single address is Port Address Translation (PAT) using NAT Overload.