Networking and Internet protocols
This article discusses the OSI Reference model and the network and Internet protocols available for connecting networks. It describes the features and services of the different protocols used in networks and on the Internet.
The Open Systems Interconnection (OSI) model was developed by the International Organization for Standardization (ISO) to provide a structural framework for designing data communication protocols. Implementing this model enabled computers to communicate with each other using different types of network protocols.
The OSI model contains seven layers:
- data link
picture box begins
picture box endsFigure 1: The OSI Reference Model
The model's upper layers deal with application issues, and the lower layers handle data transmission issues. The highest layer – the application layer – is closest to the computer user, and the lowest layer – the physical layer – is closest to the network hardware.
Each layer of the OSI model serves a particular purpose:
Table 1: OSI Reference Model
table box beginsLayer No.
This layer acts as an interface between the network and the user applications.
This layer performs data translation to and from user-recognized languages. It also performs data compression, expansion, encryption, and decryption.
This layer negotiates and maintains network connections and connections between applications. It coordinates sessions between applications at both ends.
This layer controls data transfers over the network.
This layer translates the addresses on the data packets and routes the data packets across the network.
This layer encodes and decodes the data packets into bits. It manages and handles errors in the physical layer, data flow control, and data frame synchronization. It is divided into two sublayers – the Media Access Control (MAC) layer, controlling the network access and data transfer permissions, and the Logical Link Control (LLC) layer, controlling frame synchronization, error checking, and flow control.
This layer performs data transfer through the network at the electrical and mechanical level. It governs the hardware's sending and receiving of data on a network, including defining NICs, cables, and connectors.
The services provided by each layer are implemented through a set of protocols. The list of protocols used by a system, one protocol per layer, is termed as a protocol stack. The most popular protocol stacks used in networks and on the Internet are
Transmission Control Protocol/Internet Protocol (TCP/IP)
Internet Packet eXchange/Sequenced Packet eXchange (IPX/SPX)
NetBIOS Extended User Interface (NetBEUI)
TCP/IP is the protocol suite that implements the protocol stack to be used over most networks and primarily over the Internet. The two main protocols in this suite are TCP, the transport protocol, and IP, the network protocol, in the OSI model. The TCP/IP protocol stack is included and installed in most operating systems by default. This protocol is the primary protocol on the Internet.
IPX/SPX is a proprietary protocol stack developed and maintained by Novell. IPX is a connectionless network and transport layer protocol that defines addressing and routing of data packets from one NetWare network to another. SPX is a connection-oriented extension of IPX that implements the transport layer in the OSI model.
NetBIOS (Network Basic Input/Output System) is a session-layer transport used by the operating system for connecting to network resources. NetBIOS is a set of services provided to applications, and NetBEUI is the protocol used for implementing those services. NetBEUI is a nonroutable protocol working as a network and a transport layer protocol in the OSI model. Because NetBEUI cannot route data packets, it can communicate only with devices on the same network segment or on segments connected to each other by bridges. Therefore, NetBEUI is used in small-sized and medium-sized Windows networks only.
Configuring TCP/IP involves modifying the settings of three parameters:
- IP address
- subnet mask
- default gateway
To address each machine uniquely for communication on the network, every host and router is provided with a unique IP address. All IP addresses are 32-bits long and are divided into four octets of eight bits each. The network addresses are written in dotted decimal notation with each of the eight bits written in decimal form. Each octet ranges between 0 and 255. The lowest IP address is 0.0.0.0 and the highest is 255.255.255.255.
A subnet mask is used to determine which subnetwork (subnet) an IP address belongs to, or where the network part of an IP address ends, and the number uniquely identifying a computer on the network begins.
If there is a 255 in an octet of a subnet mask, then the corresponding octet in the IP address is a part of the network address. For example, an IP address of 184.108.40.206 with a 255.255.255.0 subnet mask indicates host number 52 on the subnet 220.127.116.11.
Table 2: Identifying the Network Number and Host Number of an IP Address
table box begins
<!--table box ends-->A subnet mask also helps determine whether the source and destination machines are located on the same or different networks by comparing the network addresses of both the machines. If the network addresses are the same, both machines are on the same network. Otherwise the data has to be passed to a router for forwarding to the destination machine's router. For example, assume that Host A's IP address is 192.168.25.17 with a subnet mask of 255.255.255.0 and Host B's IP address is 192.168.25.79 with a subnet mask of 255.255.255.0. The network address for both the machines is the same, 192.168.25, which means that both the machines are on the same network.
The router, which allows access from a local subnet to the Internet, or the rest of the intranet, is termed as the gateway because it is the point where data enters and exits from the network.
To configure IP addresses, you need to understand
- Domain Name System (DNS)
- Dynamic Host Configuration Protocol (DHCP)
- Windows Internet Naming Service (WINS)
- Network link (NWLink)
Users normally address network resources by friendly text strings, such as domain names, known as fully qualified domain names (FQDN), rather than by IP addresses. With thousands of workstations connected to the Internet, it would be difficult to maintain this string-to-IP address mapping and also ensure host name conflict resolutions. To solve this problem, DNS is used to translate network resource FQDNs to IP addresses.
DNS is a group of protocols and services widely used over the Internet and TCP/IP-based networks. It is an application layer protocol and uses a hierarchical, domain-based naming scheme and a distributed database system for implementing this naming scheme.
DHCP allocates dynamic addresses to systems on a network during boot-up. It enables auto assignment of IP addresses. For static assignment, you need to manually configure your computer's IP address. When you enter a specific IP address and choose not to use DHCP, that computer uses a static IP, which does not change upon reboot.
The WINS tool is used to resolve Windows machine names – also called NetBIOS names – with IP addresses. As client machines boot, their names are registered in the WINS database. This WINS database is queried every time a computer name needs to be resolved to its IP address. When the WINS client is shut down, its name and IP address are automatically deregistered from the WINS database. Currently, WINS is being phased out by Microsoft in favor of dynamic DNS.
The NWLink IPX/SPX protocol is Microsoft's implementation of Novell NetWare's IPX/SPX protocol. Using this, Windows users can access client and server applications running on Novell NetWare servers and also access intranets running on the IPX/SPX protocol.
The Internet protocols are application layer protocols that enable machines to send data to each other over the Internet. A few of the popular Internet protocols are
- Simple Mail Transfer Protocol (SMTP)
- Internet Message Access Protocol (IMAP or IMAP4)
- Hypertext Transfer Protocol (HTTP)
- Hypertext Transfer Protocol Secure (HTTPS)
- Secure Sockets Layer (SSL)
- File Transfer Protocol (FTP)
SMTP is the preferred standard for e-mail transmissions over the Internet. SMTP is a text-based protocol, where the sender specifies one or more recipients of a message. The SMTP at the client end first verifies whether the recipient e-mail IDs exist on the mail server and only then transfers the message text. The SMTP protocol uses TCP port 25.
IMAP or IMAP4 is used to access e-mail messages on a remote server or on the Internet-based e-mail. IMAP and POP3 (Post Office Protocol version 3) are the two most common Internet standard protocols used to retrieve e-mail messages.
IMAP is mostly used in large networks and in networks where the users have to download several messages or large-sized messages. The user interface in IMAP remains active as long as the user is connected, and this enables faster access and message downloads than most other protocols. IMAP enables multiple users to connect simultaneously to the mailbox and also provides mechanisms for users to detect the changes made to the mailbox by other concurrently connected clients. IMAP also supports server-side searches for messages meeting particular criteria without requiring the user to download all the messages for performing the search.
HTTP is used to access or convey information on the Internet. HTTP primarily works as a request and response protocol between clients and servers. TCP is normally used as the transport protocol but is not formally required by the standard. The default port used for HTTP requests is 80.
HTTP defines eight methods indicating the action to be performed on the required resource:
The HTTPS protocol uses the same syntax as HTTP, but it accesses resources over port 443 and there is an additional layer of encryption and authentication between HTTP and TCP. This system is used for transferring security-sensitive data over the Internet.
The level of protection offered by HTTPS depends on the software used at the server end, the encryption algorithms used, and a correct implementation by the web browser. HTTPS ensures only that the sensitive data is protected in transit from the client's machine to the server; it does not guarantee the security of data on the server.
The SSL protocol is a cryptographic protocol that provides secure communication service for web browsing, e-mail transmissions, and other sensitive data transfers. The Transport Layer Security (TLS) is a newer version of SSL.
Telnet – short for teletype network – is a network protocol used on the Internet or on a LAN. Telnet client programs are available for all platforms, including UNIX.
Most network equipment and operating systems supporting the TCP/IP stack provide support for the Telnet service server. Telnet is a client-server protocol based on a connection-oriented transport.
Telnet is not recommended on modern systems because a lot of vulnerabilities have been discovered in its daemons and it does not encrypt data sent over its connection, making it less secure against eavesdropping.
Telnet is still used in debugging SMTP or HTTP server network services, as it is a simple way to send commands to the server and check the responses. Telnet can also function as a basic Internet Relay Chat (IRC) client.
FTP is an open standard protocol linking two computers over a TCP/IP network to transfer files from one computer to the other. Virtually all platforms support the FTP protocol.
An FTP server has file resources, runs FTP server software, and listens over the network for connection requests from other computers. The FTP client makes the file resource request, runs the FTP client software, and initiates connection to the server. Once the connection is established, the client can execute a number of file manipulation commands that include uploading, downloading, renaming, or deleting files from the server.
FTP is an insecure method of transferring files because all data, including passwords, is sent as clear text that can be viewed by anyone eavesdropping on the same network.
An Internet service provider (ISP) provides access to the Internet and Internet-related services. The earlier ISPs were phone companies because they already had a backbone of cables in place and it was easier for them to provide connectivity for any potential user. However, now an ISP can be anyone having sufficient money and expertise. ISPs provide a host of services including domain name registration and hosting, and web hosting, as well as basic Internet access.
ISPs provide direct access to the Internet to paying customers or they themselves pay and connect to an ISP upstream and use this connection to send and receive data from parts of the Internet beyond their network.
HTML is a markup language that describes how a document is formatted and presented over the Web. HTML is preferred over WYSIWYG (What You See Is What You Get) word processors, because it just informs the browser how the text is to be formatted on the screen. The interpretation of the markup is up to the browser processing the page. Therefore, it becomes possible for the web browser to reformat the page depending on the screen display configuration of the end user.
The OSI model contains seven layers. These layers are application, presentation, session, transport, network, data link, and physical. Each of the seven layers in the model provides a specific service to the layer above it.
Protocols are used to implement the services required at each layer level. The most common network protocols used are TCP/IP, IPX/SPX, and NetBEUI. TCP/IP is primarily used over LANs and the Internet, IPX/SPX is used over LANs for Novell networks, and NetBEUI is used in small-sized to medium-sized networks. Various Internet protocols are available for supporting and enabling applications on the Internet.
toc box beginsTable of Contents
| Introduction |
toc box end
copyright box begins