Book Image


By : James Denton
Book Image


By: James Denton

Overview of this book

Table of Contents (17 chapters)
Learning OpenStack Networking (Neutron)
About the Author
About the Reviewers

Network Address Translation

Network Address Translation (NAT) is a networking concept that was developed in the early 1990s in response to the rapid depletion of IP addresses throughout the world. Prior to NAT, every host connected to the Internet had a unique IP address.

OpenStack routers support two types of NAT:

  • one-to-one

  • many-to-one

A one-to-one NAT is a method in which one IP address is directly mapped to another. Commonly referred to as a static NAT, a one-to-one NAT is often used to map a unique public address to a privately addressed host. Floating IPs utilize one-to-one NAT concepts.

A many-to-one NAT is a method in which multiple addresses are mapped to a single address. A many-to-one NAT employs the use of port address translation (PAT). Neutron uses PAT to provide external access to instances behind the router when floating IPs are not assigned.

For more information on network address translation, please visit Wikipedia at