Specifications
Table Of Contents
- About This Document
- Understanding Networking and IP Addressing
- Introduction to Networking
- Networking using IP
- Niagara Considerations
- Additional Information
- Configuration and Troubleshooting Tools
- Connecting on a LAN
- Connecting with Direct Dial
- Connecting to an ISP
- Using Security Technologies
- Configuration Files Used for Communication
- Glossary
- Index

Niagara Release 2.3
Revised: May 22, 2002 Niagara Networking & Connectivity Guide
Chapter 1 Understanding Networking and IP Addressing
Networking using IP
1–18
that provides access to a remote host. And, DNS (Domain Name System) and WINS
(Windows Internet Name System) servers resolve names into actual IP addresses,
which are required to identify resources on TCP/IP networks.
Presentation and Session Layers
Layers 6 and 5 of the OSI model are not formally defined in TCP/IP. The services are
indeed performed, if needed, in other layers of the TCP/IP stack.
Transport Layer
TCP establishes a virtual connection between two stations before they transmit data.
Once the connection is established, both sides negotiate the maximum size of a data
packet. Although TCP supports packet sizes up to 64KB, in most cases the actual
packet size is determined by the underlying network (i.e., Ethernet, which supports a
maximum frame size of 1518 bytes). TCP attaches a header to the packet that
contains the source and destination ports as well as a sequence number, then it hands
the packet over to the network layer (IP).
TCP uses a sliding window system that adjusts the number of packets to be received
before an acknowledgement is sent based on network traffic. This flow control
mechanism is a real-time means of reducing transmission errors resulting from
excessive collisions.
Network Layer
The IP protocol, running on the network layer, accepts data packets from TCP (or
UDP) above it and prepares them for the data link protocol layer below. IP converts
the IP address of the target station into a physical station address (MAC address) and
fragments the packets (if needed) into the required frame size. IP uses the ARP
(Address Resolution Protocol) to broadcast the request onto the network, which
causes the target station to respond with its MAC address. IP then outputs the packets
(called datagrams) and hands them over to the data link layer below it along with the
MAC address of the target station or router.
Data Link Layer
Ethernet is the most widely used data link layer protocol. It receives the datagram
from the network layer above it and wraps it in its own frame format, which includes
a header with source and destination MAC addresses and a trailer that contains
checksum data. Ethernet then broadcasts the frame onto the wire using the
CSMA/CD (carrier sense multiple access/collision detection) collision avoidance
methodology.
Physical Layer
The physical layer specifically addresses the hardware used to transmit data over the
network. This includes the voltage used, the timing of data transmission signals, the
requirements for establishing the initial communications connection, connectors, and
interfaces to data transmission media. This layer provides the electromechanical
interface through which data moves among devices on the network. It specifies the