Yale University Library

 

Library Access Integration Services

Quick Links
Network Protocols

A network protocol is a set of rules and conventions for sending information over a network.  These rules govern the following parts of messages exchanged among network devices:

  • Content
  • Format
  • Timing
  • Sequencing
  • Error control

Protocols we use on Library workstations:

NetBEUI A non routable protocol designed for use in small LANs.
TCP/IP Transport Control Protocol/Internet Protocol - The default wide area network protocol that provides communication across diverse interconnected networks.
NWLink Enables Novell NetBIOS packets to be sent between a Netware server running Novell NetBIOS and a Windows computer or between two Windows computers.
NWLink IPX/SPX
compatible transport
An implementation of IPX/SPX protocols which is used by Netware networks.  If you need to attach drives on a Novell server you will need this.
DLC Protocol Data Link Control - Enables computers to connect to an IBM mainframe.  Enables the setup of printers attached directly to the network, such as the HP 5N that has its own network board and is plugged directly into the network rather than attached to a computer then shared.
Appletalk The protocol though which Macintosh computer communicate.  

Open Systems vs. TCP/IP

Open Systems

A set of protocols that would allow any two different systems to communicate regardless of architecture is called an open system.  The Open Systems Interconnection model (OSI Model) is a seven layer architecture for allowing various devices to communicate.  If this were fully developed it would allow any two computers to communicate as long as they were connected.

Each layer in the OSI Model performs separate functions and communicates with the layers above and below it.  Higher layers deal more with user services and applications, the lower layers deal with the actual transmission of information.

TCP/IP

TCP/IP is the standard protocol for communications on the Internet.  It is actually two protocols.  Transmission control protocol and Internet protocol.  These two protocols correspond roughly to layers 3 and 4 of the OSI model although they are not part of the OSI model. 

To communicate on the Internet you must have:

  • A LAN that has a router connected to the internet or an internet service provider.
  • A computer with a network board.
  • A unique IP address.

Large organizations on the Internet must have an intermediate level of network organization and expertise.  Routers might be configured to connect several departmental LANs in several buildings. All traffic outside the organization would typically be routed to a single connection to a regional network provider.  In the case of Yale it has three T1 lines that connect to MCI our internet provider in New Jersey. 

A Computer needs three thing to be able to communicate on the Internet:

  • IP address (this should be the only machine with this address) : 130.132.80.104
  • Subnet mask: 255.255.255.0
  • Router address (every computer using this router must have this address): 130.132.80.1

All computers having an IP address starting with 130.132.80.xxx are on the same department LAN, and have messages sent through it.  Any communications with machines having IP addresses beginning with an address different than 130.132.80.xxx will be sent through the router 130.132.80.1.

TCP/IP allows you to use the following applications:

  • Simple Mail Transfer Protocol (SMTP) - allows to send electronic mail (EMail) to thousands of other computers all over the world.
  • File Transfer Protocol (FTP) - allows sending and receiving files from one system to the other under user command.
  • Telnet - provides the capability to log on to a remote computer system. The user then is able to act as if directly connected to that system.
  • Web Browsing - is a hierarchical hypertext system designed to allow network users to navigate the Internet.

To determine the IP information on your computer you can go to the Command Prompt and type: IPCONFIG.  Your computer will return your IP address, Subnet mask and Default gateway.  You can also get this information in the network setup.

How an IP Address is assigned

There are 2 ways of assigning an IP Address to a workstation:

Static IP Address One where the address is assigned to that machine and is hard coded into the TCP/IP configuration.  The address must be assigned by a network administrator.
DHCP Dynamic Host Configuration Protocol A way of centrally and automatically assigning assigning IP addresses.  Each time a DHCP client starts it requests an IP address, Subnet mask and the router address from the DHCP server.  When the request is received by the DHCP server an address is selected from the pool of address in its database.

In the future the University may be using DHCP campus wide instead of assigning fixed IP addresses.  What absoluteley must have a fixed IP address is a DHCP server.  It must always have a specific address so its clients know where to go to get and address from.  DHCP address can be assigned to specific machines by having the hardware address (MAC Address) of the clients entered into the DHCP database.

The MAC address is the media access control number. It uniquely identifies a piece of network communications hardware on the internet.

Domain Name Servers

A domain name server is a protocol that translates a symbolic internet address to a 32-bit numeric address.  It is the unique name that identifies an Internet site. Domain Names always have 2 or more parts, separated by dots. The part on the left is the most specific, and the part on the right is the most general. A given machine may have more than one Domain Name but a given Domain Name points to only one machine or group of machines. For example, the domain names:

yale.edu

mail.yale.edu

can all refer to the same machine, but each domain name can refer to no more than one machine.

Usually, all of the machines on a given Network will say the same thing in the right-hand portion of their Domain Names (yale.edu in the examples above). It is also possible for a Domain Name to exist but not be connected to an actual machine. This is often done so that a group or business can have an Internet e-mail address without having to establish a real Internet site. In these cases, some real Internet machine must handle the mail on behalf of the listed Domain Name.

http://www.internic.net

Name     Descriptions
.com Comercial organizations, such as microsoft.com
.gov Government organizations, such as nasa.gov
.mil Militrary organizations, such as army.mil
.net Networking and Internet Service Providers PSI.net
.org Non-comercial or non-profit angenices
.int International organizations
.edu educational institutions

 BackForward

Revised 7/23/2001