Computer Network

COMPUTER NETWORK Definition: Inter-connection between two or more computers via a transmission media for faster and easier communication and resource sharing

Objectives, Uses & Benefits: • Faster and easier communication – i. No need for removable devices (floppies & pen drives) ii. No need for multiple printers • Resource sharing – Sharing hard drive

•

•

A department of 10 users, each requiring 25 GB of hard disk space, can share a single hard disk of 250 GB centrally located instead of allocating each user with dedicated 25 GB hard disk Significant reduction in the per unit cost of storage

Sharing data

• •

Sharing a document or an image file or a corporate database system Multiple users can access the same data at the same time Sharing peripherals

•

Sharing devices like – printer, scanner, CD-ROM Sharing Applications (softwares)

• • •

Electronic Mails It consists of a mail program that allows the user to send and receive messages It provides a messaging system to support store-and-forward functions. The system contains a dedicates computer called mail server that acts as the central repository for incoming & outgoing messages

•

• •

•

The central repository stores the messages submitted by a user and waits until the other user logs on and then sends him the messages. The users can send and read messages by using clientmessaging software provided by the messaging software. Some of the commonly used network messaging software is Microsoft Exchange and Mail, Lotus CCMail and Notes and Novell GroupWise.

Fault tolerance (resilience) – i. Effective system of backing up data (centralized data backup system) ii. RAID Technology (HD Mirroring) iii. Replicating data using backup servers iv. Cluster Technology (multiple backup servers)

• Administration – managing security i. User authentication method (username & password) ii. IP address validation method (access-list technique) iii. Virus security policy (downloading & updating antivirus softwares) NETWORKING COMPONENTS: 1. Resource – Anything related to data and communication is a Resource

Computer Resources: • Printers • Scanners • Storage devices (e.g. HDD, FDD, CD-ROM, etc) • Fax devices • Electronic mails • Modems • Files, folders & documents Types of Resources: • Local Resources • Remote Resources • Resource Accessibility: • Locally accessible resource • Remotely accessible resource

2. Transmission Media – Any medium to exchange data or information

Types of Media: • Wired Media / Bounded Media • Wireless Media / Unbounded Media 3. Server – A server is a remote computer that serves information on request. It receives the queries, processes them and then replies to the queries. It has the capability to handle multiple connections from different sources. When a request is sent to the server, it either answers to the request and updates the information, or it sends the request to the right destination. Some popular types of servers are: • Print server • File server • Database server • Mail server • FTP server • Web server A print server allows the users on a network to share printers A file server allows the users on a network to share remote files, folders and storage devices A database server contains a single common database that the users can query and edit from remote locations A mail server processes and delivers e-mails in an organization having intranet A FTP (File Transfer Protocol) server manages file transfers A web server hosts web pages to be accessed from anywhere in the world

Classification of server: • Dedicated server: A server in which local resources are not accessible once the server program is invoked on it •

Non-dedicated server: A server in which local resources are accessible while the server application keeps running in the background. Such servers may hang up due to faulty program handling or execution by the user, causing the whole network to crash.

Note: A network computer can be termed as a server only when a server application is running on it that listens for requests and delivers services for those requests 4. Client – A client is a computer on the network on which a client application is running that requests for and receives some services from the server 5.

Node – Any computer or any network resource is called a NODE

6. Workstation – A network computer that is fine-tuned to run one particular application. A workstation may be a node, but all nodes may not be workstations 7. Host – It is a TCP/IP terminology to define any network device that allows a client to access its resources

8. Peer – A computer on the network acting both as a client as well as a server 9. Segment – When a large network is broken down into a number of smaller networks and linked by connecting devices like hubs and switches, we call it a network segment 10. Backbone – The link between a numbers of smaller network segments to create a large network is the network backbone

NETWORKING MODELS: • Centralized • Distributed Centralized: Data is stored, organized and processed by a centralized server that is fed from remote and dumb terminals (having no hard disk and processing capability) acting only as I/O devices and share no resource. Distributed: Data is entered and processed at the remote terminals but stored into a server. The remote terminals share resources among themselves and also with the server. 95% of the data processing is done at the terminals. The server acts as the BACKEND and the terminals act as FRONTENDS. Types of Distributed Networks: • Peer-to-Peer Networking • Client/Server Networking

Peer-to-Peer Networking: • Two or more computers share their individual resources, such as, hard drives, CD-ROM drives, printers, etc. • Each computer acts both as a client and a server • Each workstation has equal capabilities and responsibilities Advantages: • Network is faster and cheaper • Easiest way to build a network • No network administrator is required • User-level security Disadvantages: • No central administration • Less security

Client/Server Networking: • A computer is dedicated as a server that provides services to multiple computers acting as clients or workstations over the network • An administrator is assigned to control the overall network and its security • Implementation of server application and client application is mandatory • All PCs depend on the central server that does the majority of work • A request from the client to the server contains the following information – • The server address • The request • The return address

Advantages: • Strong central security • Central file repository provides easy data backup • Ability of servers to pool available hardware and software, lowering overall costs • Ability to share expensive equipments, such as laser printers • Optimized dedicated servers that provide faster sharing of network resources Disadvantages: • Implementation is costly and complicated • Dedicated Network Administrator is required, which is again a costly affair • When the server is down, all the network services are unavailable

NETWORK OPERATING SYSTEMS: Client Operating Systems: • DOS • OS/2 • Windows 3.x • Windows 9x, Me • Windows NT Workstations (3.x, 4.x) • Windows 2000 Professional • Windows XP Professional • Novell Clients • UNIX Workstations • Linux Workstations

Server Operating Systems: • Novell Netware 2.x, 3.x, 4.x, 5.x • Windows NT Server 3.x, 4.x • Windows Server 2000 • Windows Server 2003 • SCO UNIX Server • Linux Server NETWORK CLASSIFICATIONS: Categorized on the basis of distance and structure – • Local Area Network (LAN) • Campus Area Network (CAN) • Metropolitan Area Network (MAN) • Wide Area Network (WAN) • Internet LAN: • A network of computers –  Within a room  Between rooms within a floor  Between floors within a building  Between buildings within a campus

• Can cover a distance of maximum 10 Kilometers • Much easier and cheaper for implementation • Data transmission speed is relatively high • Practically no error in data transmission • Mostly uses simple cabling methods for connecting computers

Uses networking devices like hubs, bridges, switches, Ethernet, token ring, FDDI, etc. • LAN Types:  Ethernet  Token Ring  Token Bus  FDDI (Fiber Distributed Data Interface) •

MAN: • A network of multiple LANs (a LAN of LANs)

Usually expands throughout a city (intra-city) – connects different offices in the same city • Can cover a distance of maximum 100 Kilometers • Relatively expensive for implementation than LAN • Slower data transmission than LAN • Connects diverse devices using optical services like SONET, SDH and ATM SONET = Synchronous Optical Network Standard SDH = Synchronous Digital Hierarchy Standard; ATM = Asynchronous Transfer Mode •

WAN: • A network of computers that expands throughout the world Connections are made over lease lines, satellite links or microwave transceivers • Most expensive communicating devices are used • Implementation and configuration are very complicated •

INTRANET • Network that is limited to a single organization or company • It is normally implemented as a LAN or WAN network INTERNETWORK • When a number of smaller network segments are interconnected using some networking devices such as ROUTERS to form a large network, we call it Internetwork • Internet is the perfect example of Internetwork

Case Study – I MoneyMaker is a bank having its registered office at Delhi. It has branches at Mumbai, Chennai, Hyderabad and Bangalore. The operating departments in the bank are Finance, Insurance, Loan, IT, Marketing, Customer Service and HR. MoneyMaker bank uses LAN as their computer network for each department. All the branches of the bank from different cities are connected through WAN. The bank is expanding and decided to open its branches at different locations in the city.

Problem Determine which type of network to be used within a city

Suggested Solution Use LAN computer network for each department in the new branch. MAN can be used for connecting the different branches of the bank within the city

NETWORK TOPOLOGIES: • Geographical orientation and arrangement of networking nodes • Two type of Physical Topologies are available – point-to-point and multipoint Point-to-Point • A one to one or back to back communication where only two devices share a connection • A specific level of bandwidth is assured to the user as there are only two devices that share the connection

Multipoint • Three or more computers share the connection • The bandwidth available for each user depends upon the number of computers on the network • Bandwidth changes drastically depending on the network load Types of Multipoint Physical Topology – a. Bus b. Star c. Ring d. Mesh e. Hybrid i. Star-Bus ii. Bus-Ring iii. Star-Ring iv. Bus-Star-Ring

Bus Topology

Star Topology

Ring Topology

Mesh topology

Bus Topology – Features: • Consists of a continuous length of backbone cable called Trunk or Bus • The trunk used is usually of Coaxial cable type • The cable has a 50 Ohms terminating resistor called BNC Terminator attached at each end • The nodes are attached to the trunk either directly in a linear method or by using a drop cable • The trunk or the drop cables are attached to the nodes at their Network Adapter Cards / Network Interface Cards / LAN cards • All nodes on the bus topology have equal access to the trunk

• Only one device can transmit data at a time over a bus topology • The data travels along the bus in both directions until a node picks it up • If the message is not recognized and collected by any of the nodes, it reaches the end of the bus and dissipates at the terminator • Without termination, when the signal reaches the end of the wire, it bounces back and travels up the entire length of the bus • When a signal echoes back and forth along an open-ended bus, it is called ringing • Continuous ringing would cause other nodes to stop transmission and the total network would collapse in no time

Bus • • •

Topology – Advantages: Simplest, cheapest and easiest to implement and to maintain Reliable in a very small network Requires least amount of cables to connect the computers • Easy to extend a bus by joining two cables using BNC Barrel connector or by using a repeater

Bus Topology – Disadvantages: • Relatively difficult to re-configure and trouble-shoot • The signal quality deteriorates with increase in the cable length as well as the number of nodes • Since one device can transmit at a time, network media access is very slow Ring Topology – Features: • A circular loop of point-to-point links • Each device is connected either directly to the ring or indirectly through an interface device or drop cable • The ring may consists of either Coaxial or Fiber-Optic cable • Message travels around the ring from node to node in a very organized manner and in a definite direction • Each node checks the message for a matching destination address • If the address doesn’t match, the node simply regenerates the message and sends it on its way • If the address matches, the node accepts the message and sends a reply to the originating sender • The unidirectional ring topology may break down due to cable failure at any point • Sometimes dual counter rotating rings are used to avoid the breakage in a single ring and subsequent failure in networking

• Dual counter loops use two physically separated rings operating in opposite directions

Ring Topology – Advantages: • Initial installation procedure is very simple • Least amount of cable is required after Bus Topology • No terminator is required as the cables are laid in ring fashion Ring Topology – Disadvantages: • Difficult to expand and re-configure • Adding or removing a device requires re-configuration of the entire network • Any break in the loop affects all devices on the network Star Topology – Features: • Consists of a central controlling unit with dedicated point-to-point links with the networking devices • This layout keeps the line of communication open and free of traffic • All the nodes have equal access to the network and can communicate at the same time • When message is sent from any node, the central device receives the signal • The central device resends the signal either to all the computers or only to the destination computer, depending on the addressing scheme and the central device used Star Topology – Advantages: • Easy to remove and add new devices on the network • Easy to diagnose network faults • Problematic media and devices are automatically detected and isolated Star Topology – Disadvantages: • Somewhat difficult to install and needs expertise to some extent • Requires a great deal of media and cables • Costly affair than Bus or Ring topology Mesh Topology – Features: • The only true point-to-point design that uses dedicated link between every device on the network • Most commonly used topology in WAN links • More practical approach is hybrid topology Mesh Topology – Advantages: • Easy to troubleshoot as each link is independent of all others

•

Faults are easily identifiable and isolated

Mesh Topology – Disadvantages: • Excessive wastage of transmission media due to dedicated pointto-point links between every device • Very difficult to install and reconfigure • Each communicating device requires multiple LAN cards and hence costly affair Hybrid Topology: Most practical hierarchical topology used in modern day networking

ISO – OSI REFERENCE MODEL ISO – International Standards International Standardization

Organization

/

Organization

for

OSI – Open Systems Interconnect / Interconnection • Primary architectural model for networks • A conceptual blueprint of how communications should take place • Depicts the guidelines for network data transmission between computers that have different hardware vendors, software, operating systems and protocols • A set of guidelines providing a framework for creating and implementing network standards, devices and networking schemes • Refers to the common points for networking between dissimilar systems • Divides the processes required for effective communication into logical grouping called layers • Breaks the communication into 7 logical layers: 7. Application Layer 6. Presentation Layer 5. Session Layer 4. Transport Layer 3. Networ Layer 2. Data Link Layer 1. Physical Layer

•

(All People Seem To Need Data Processing / Please Do Not Throw Sausage Pizza Away) • Each layer performs its role independently but in coordination with the overall goal of communication

Peer Entities: Layers at the same level in two different machines on the network are called Peer Entities. For example, the Application Layer on one machine is the Peer Entity of the Application Layer on the other machine. Peer Communication

Protocol: Well-defined set of rules and conventions governing the format of messages required for the peer entities on the network to communicate with each other is called the PROTOCOL. • Peer entities talk to each other using the same protocol • Each layer on the network has its own unique protocol Data Flow: • Encapsulation • Decapsulation

Encapsulation & Decapsulation: Communications flow within the OSI Model vertically – meaning that each layer communicate to the layers just above and below it. The user data is intercepted at the topmost layer – the Application Layer and

travels down the model layer to layer. At each of the layers the user data is formatted with specific headers unique to the layer. A header’s primary job is to carry out the function of its layer. The combination of data and header at a specific layer acts as the data for the layer just below it. The header is being added at each layer until the user data reaches the lower most layers – the Physical Layer. This process of adding headers to the actual data, one by one, by all the layers in succession, is called Encapsulation. Once the data is received by the Physical layer on the receiving side, it starts traveling vertically upwards along the seven layers. But this time each layer strips off the header information which had been added by its peer at the sending side. The header is being stripped off at each layer until the user data reaches the upper most layers – the Application Layer, and the actual data is made available to the user. This process of stripping off headers, one by one, by all the layers in succession, is called Decapsulation. OSI Model at a Glance: OSI Layers Application

Hierarch y Upper layer

Data Format Message

Presentation Upper layer

Message

Session

Upper layer

Message

Transport

Middle layer

Segment

Network

Middle layer

Packet

Data Link

Lower layer

Frame

Physical

Lower layer

Bits

Functions Services – interfaces user applications Services – negotiates data exchange Services – establishes connections Networking – end-to-end connection Networking – routes data Communication – accesses n/w media Communication – transmits data

Application Layer • Provides a user interface • Provides file, printer, message, database & application services • Defines how interaction occurs between network services (applications) and the network • Keeps data in the form of messages • Services – file transfer services (TELNET, FTP), messaging services (SNMP, NNTP), web browsers (HTTP), e-mails (SMTP, POP3, IMAP4), word processing, etc. Presentation Layer • Provides appropriate data presentation • Provides data encryption, compression and translation services • Keeps data in the form of messages • Services – ASCII, JPEG, MPEG, GIF, TIFF, RTF, BMP, MIDI, etc. Session Layer • Provides end-to-end connection • Establishes, synchronizes, maintains & terminates a connection (session) • Practical functions such as security authentication, connection ID establishment, data transfer, acknowledgements and connection release take place • Provides dialog control – keeps different application data separate • Keeps data in the form of messages • Services – Remote Procedure calls (RPC) & Network File System (NFS) Transport Layer • Provides reliable and unreliable delivery (flow control) • Performs error correction before transmission • Keeps data in the form of segments • Multiplexing connections • Protocols – TCP, UDP, SPX Network Layer • Provides routing for packets • Provides logical addressing • Keeps data in the form of packets • Protocols – IP, IPX, DECnet, AppleTalk, GRP, IGRP, EIGRP, OSPF, etc.

Data Link Layer • Provides access to media using MAC address • Performs error detection but no error correction • Combines packets into bytes and bytes into frames • Keeps data in the form of frames Physical Layer • Moves bits between devices • Specifies the type of cable, pin-out of cables, type of connectors, type of interfaces • Defines the physical topology • Keeps data in the form of bits

LAYER – 1: THE PHYSICAL LAYER FUNCTIONS: • Bottom-most layer of the OSI Model • Deals with physical and measurable entities that follow the low level protocols for transmitting data • Encodes data into bits and decodes bits into data (converts the outgoing data bits into electrical signals & incoming signals into binary data) • Communicates directly with various types of actual communication media • Specifies the electrical, mechanical, procedural and functional requirements for activating, maintaining and de-activating a physical link between end systems • Specifies the type of cable, pin-out of details, type of connectors and the type of interfaces • Defines the physical topology PHYSICAL FEATURES: 1. Electrical Properties 2. Data Signaling and Encoding 3. Physical Topology 4. Transmission Media 5. Transmission Devices 6. Data Synchronization 7. Data Bandwidth Electrical Properties: 1. Resistance – Electrical resistance is the result of friction offered by the material of the cable that carries the signal. Thicker wires have low resistance than the thinner wires. Resistance results in the drop of voltage and the subsequent loss of data. So the cables used must not be thin enough for the electrical signals to dissipate due to high resistance. 2. Capacitance – The networking cables may charge up to high

capacitance when the signal is at a very high voltage and get discharged when the signal is at a very low voltage. This causes a very slow change in the state of signal (from high to low and from low to high), resulting in signal distortion and data errors. LAN components must be designed to have as little capacitance as possible.

3. Impedance – Since high frequency AC signals travels on the

wire’s surface, drastic changes in the size of the cabling wires must be avoided that would cause impedance to the flow of signals. 4. Noise – Noise is another electrical property that can cause the

square waves of communication to become fuzzy which alters the signal strength to create data errors. There are 2 types of noise disturbances: Radio Frequency Interference (RFI), and Electro-Magnetic Interference (EMI) Fluorescent Lights, Arc Welders, Transformers, Lightning, etc. can create significant electrical field causing noise problems. 5. Attenuation – It is the reduction in signal strength as electrons

pass through the cables. It is the result of resistance, capacitance, impedance and other characteristics of the wire. It is measured in decibels per thousand feet (dB/KFt). Attenuation is one of the major factors that determine the maximum cable distance for most topologies. The signal must be strong enough for the receiver to separate true data from any noise picked up along the way. 6. Cross Talk – Whenever current is sent through a wire, it creates a

magnetic field. Other wires nearby will absorb some of that magnetic field that creates reverse current. This phenomenon is found in telephone conversation. Current that leaks from one pair of wires to another, also drains data along with it. Cross talk is the advanced form of noise and must be properly guarded by good insulation. 7. Propagation – Speed of data through a cable is measured in

Nominal Velocity of Propagation (NVP) which is between 0.78 and 0.95. This means that data travels through the cable at 78% to 95% of the speed of light in vacuum (light travels in vacuum at a speed of 186,000 miles/sec). Data Signaling and Encoding: There are 2 types of data signaling – Digital & Analog 1. Digital Signaling –

• Represents information as a discrete flow of voltage with respect to time • Uses the full capacity of the transmission media • Cannot travel long distances as the signal strength becomes weak • A high voltage represents the bit value as 1

• A low voltage represents the bit value as 0 • Hence, a digital signal has one of two states – either high/1 or low/o • As the signal jumps from one level to another, it is measured during a time slot called Bit Interval • The presence or absence of a pulse determines the bit value at each interval • There are 2 types of Digital Encoding methods – Current State & State Transition i.

Current State –  Measures the presence or absence of a voltage level during a bit interval  The sending and the receiving devices must be synchronized to use the same bit interval

ii.

State Transition –  Measures the transition between two states during a bit interval  No transition represents a zero, and any –ve or +ve transition represents a one  More reliable method of data encoding as the transition allows the sender and the receiver to stay in sync all the time

2. Analog Signaling –

• Represents information as a continuous and variable flow of voltage with respect to time • Does not use the full capacity of the transmission media; hence multiple signals can be transmitted over a single transmission media • Can travel a long distance without deteriorating the quality of the signal • Signals are smooth and rhythmic, hence an ideal platform for data communication • Analog waves are measured according to its features - Amplitude, Frequency and Phase • There are 3 types of Analog Encoding methods – iii. iv. v.

Amplitude Shift Keying (ASK) Frequency Shift Keying (FSK) Phase Shift Keying (PSK)

Physical Topologies: • Point-to-Point

• Multipoint – Bus, Star, Ring, Mesh, Hybrid

Transmission Media: • Media is anything that carries a message • Transmission media are the elements that establish the physical link over which signals are exchanged between the transmitting and the receiving stations Types of Transmission Media: A. Bounded (Wired) Media 1. Coaxial Cable a) Thick Coaxial b) Thin Coaxial 2. Twisted Pair Cable a) Shielded Twisted Pair b) Unshielded Twisted Pair 3. Fiber Optic Cable a) Mono Mode b) Multi Mode B. Unbounded (Wireless) Media 1. Micro Wave a) Terrestrial b) Satellite 2. Infrared a) Unidirectional (Point-to-Point) b) Omni directional (Broadcast) c) Reflective 3. Radio Wave a) Low-power Single Frequency b) High-power Single Frequency c) Spread Spectrum Transmission Devices: Provides basic data communication functions – a) Translation (frames into bits & vice-versa) b) Transformation (digital signal into analog signal & vice-versa) c) Transmission (one networking device to another) Types of Data Transmission Devices – A. Communication Devices 1. Modems • Transforms digital signal into analog MODULATION • Transforms analog signal into digital DEMODULATION

signal

–

signal

-

2. Multiplexers • Multiple input signals are converged (funneled) and transmitted over a single transmission media – MULTIPLEXING • The converged signals are separated and received as multiple channels as output – DEMULTIPLEXING Multiplexing is a technique in communication that allows multiple messages or signals to share a transmission channel. A device that performs multiplexing is called a multiplexer. The input channels are normally low-speed, while the output channel is high-speed, with enough bandwidth to accommodate the multiple slow channels. 3. CSU/DSU Such devices are used if the transmission media is also digital (leased line & T-1 circuit) Consists of 2 parts – a) CSU – Channel Service Unit Line management (grounding & loop-back testing) b) B.

1.

2.

3. 4.

5.

DSU – Data Service Unit

Networking Devices Network Interface Cards (NICs) a) Ethernet b) Token Ring c) Fiber Distributed Data Interface (FDDI) Media Connectors a) BNC b) RJ-45 / RJ-11 c) ST / SC Repeaters a) Amplifiers b) Signal Regenerator Hubs a) Passive Hubs b) Active Hubs c) Intelligent Hubs d) Stackable Hubs Multi-Station Access Units (MSAU)

Data Synchronization: Data synchronization is the process of establishing consistency among data on remote sources and the continuous harmonization of the data over time.

Data Bandwidth:

LAYER – 2: THE DATA LINK LAYER FUNCTIONS:

• •

• • •

Receives raw data bits from the Physical Layer Transforms the data bits into organized structure called frames Adds control information to the data frame (Preamble – 7bytes, Start Frame Delimiter/Synch – 1byte, DA – 6bytes, SA – 6bytes, Network Layer Protocol Identifier – 2bytes, Data Length – variable, Frame Check Sequence/CRC – variable) – [SHOW DIAGRAM] Uses MAC address Performs error detection but no error correction

SUB-LAYERS: [SHOW DIAGRAM] 2 Sub-layers by IEEE – • Media Access Control (MAC) Sub-layer • Logical Link Control (LLC) Sub-layer MEDIA ACCESS CONTROL (MAC) SUB-LAYER: Defines the way in which multiple NICs share a single transmission medium MAC Functions: • Physical Addressing • Media Access Method • Logical Topology Physical Addressing: • A protocol independent hardware address • Also called the MAC Address • A unique 12-digit hexadecimal (48 bit) identifier of a particular PC on a network • Regulated by IEEE Standards Organization • Hard-coded within the EEPROM of the NIC • No 2 NICs can exit with the same MAC Address • A MAC Address is divided into 2 parts – a) 1st six Hex. Digits identify the vendor of the NIC (Organizationally Unique Identifier – OUI) b) 2nd six Hex. Digits uniquely represents the NIC Each data frame contains 2 MAC Addresses – A source MAC address A destination MAC address • Again, destination MAC address can be of 3 types – [SHOW 3 DIAGRAMS] a) Unicast MAC Address b) Multicast MAC Address (creating groups) c) Broadcast MAC Address (turning all bits to 1)

a) b) c)

• • •

Media Access Methods: Provides guidance to data signal to reach destination Promotes harmonious connectivity between multiple network devices 3 methods (protocols) of Media Access are – Contention Token Passing Demand Priority Contention: A protocol that works on physical bus topology with standard Ethernet networks All transmitting devices have equal access to the network media Data may be lost or damaged due to possible collisions Collisions occur when 2 devices try to transmit at the same time Collisions result in disrupting each other’s data signaling [SHOW DIAGRAM]

Modern method to reduce the chances of possible collisions – CSMA (Carrier Sense Multiple Access) Carrier Sense – Sensing the transmitting media for any existing signal If no signal is sensed, sending station assumes the media to be free and starts transmitting In the event of any existing signal, sending station backs off and waits until the media becomes free Multiple Access – multiple stations can access the media at the same time 2 types of CSMA: • CSMA/CD – CSMA with Collision Detection • CSMA/CA – CSMA with Collision Avoidance 1.

2. 3. 4. 5.

7.

Procedure for CSMA/CD: NIC of the sending station first senses for any exiting transmission within the media If the media is free it starts transmitting data; if not, it backs off and waits for the media to be free After transmitting data, NIC listens to the media to detect any collision If a collision is detected, a collision signal is sent and other NICs stops transmitting All NICs will wait far a random delay of time before retransmitting 6. If collision occurs for the 2nd time, transmitting stations repeat the above steps but doubles up the random time-out before they again retransmit Once the transmissions are successfully done, other machines are allowed to transmit Procedure for CSMA/CA: Uses time-slicing method of media access Collision is altogether eliminated A protocol used by AppleTalk networks The transmitting device sends a Request to Send (RTS) message The network server sends a Clear to Send (CTS) message Abort signal is send after the transmission is over Logical Topology: • Defines the way data is moved through a network • Types are same as the physical topology • Information flow specifies the type of topology to use Token Passing: (page – 81/82) Demand Priority: (page – 83) LOGICAL LINK CONTROL (LLC) SUB-LAYER: Establishes and maintains link between communicating devices LLC Functions: • Flow Control • Error Checking • Provides both Connection-less & Connection-oriented Services Data Link Flow Control: • Network devices have different transmission speeds, storage capacity & processing capabilities • Flow control regulates the transmission of data between different devices • 2 types of flow control – a) Sliding Window Flow Control b) Guaranteed Rate Flow Control (Stop & Wait) Sliding Window Flow Control: A special buffer called Window is created

Window allows a certain amount of data frames to flow thru in a given period of time a) b)

• •

• • •

2 types of Window Flow Control – Static Window Flow Control Dynamic Window Flow Control

Static Window Flow Control: Buffer is fixed in size Sender must fill up the buffer to its full capacity Data frames are sequenced and then transmitted Receiver sends back an acknowledgement & requests for the next buffer This procedure is inflexible and inefficient Dynamic Window Flow Control: • Buffer size is variable as per the receiver’s capacity • This procedure is very flexible and efficient Guaranteed Rate Flow Control: • Data is transmitted at a fixed rate • Also called Stop & Wait method of Transmission • The sender and the receiver agree upon an acceptable transmission rate • No buffer is used • When the receiver’s memory is full, it suspends transmission • A signal to resume transmission is sent when the memory is free again Data Link Error Checking: • Uses a complex calculation called Cyclic Redundancy Check (CRC) • CRC ensures integrity of data frames • Before transmission, data frames are passed thru an algorithm • Algorithm is a complex mathematical calculation to solve a problem step-by-step • Algorithm generates a 16 or 32 bit unique number called Checksum • With a change in the frame contents, the CRC value (Checksum) also changes • Checksum is places in the CRC Data Link Field • The receiver performs the same computation at its end and generates a Checksum • If the CRC values match, the data is sent error-free & an acknowledgement is sent • If the CRC values don’t match, the receiver responds with a NAK Connection Services: • Connectionless Services – sending and receiving frames without any flow control or error checking or sequence control • Connection-oriented Services – provides strict rules for flow control, error checking and sequencing with proper acknowledgement NOTE: Physical Topology

Logical Topology

•

Physical arrangement of network

•

Media access technology of network

•

Feature of the Physical Layer

•

Function of the Data Link Layer

IEEE STANDARDS: IEEE: INSTITUTE OF ELECTRICAL AND ELECTRONIC ENGINEERS • A New York City – based organization • Established in 1962 • An important standards-setting body for the IT industry • Comprises of over 300,000 members, including engineers, students and scientists • IEEE 802 is a set of standards that refers to the Networking Technologies IEEE 802 Networking Standards: • A set of Physical Layer & Data Link Layer Standards • Published by IEEE in 1985 • Revised and reissued as 8802 protocols by ISO • A standard is first developed, then a product is created to conform to the set standard Standards

Networking Technologies

802.1 802.2 802.3 802.4 802.5 802.6 802.7 802.8 802.9 802.10 802.11 802.12

LAN/MAN Management & Media Access Control Bridges Logical Link Control CSMA/CD Token Bus Token Ring Distributed Queue Dual Bus (DQDB) for MAN Broadband LAN Fiber-Optic LAN & MAN Integrated Services LAN Interface (ISLI) LAN/MAN Security Wireless LAN Demand Priority Access Method

IEEE 802.2 Standards: • An LLC sub-layer standard • Uses 802 protocols such as 802.3 & 802.5 • Provides error control & flow control • A standard Network Layer service interface • Adds headers to the data frame • Specifies upper layer protocols used for multiple protocol support IEEE 802.3 (802.3u) Standards: • Specifies a network that uses bus topology, baseband signaling & CSMA/CD • Originally developed to match DIX (Digital, Intel, Xerox) Ethernet Networking technology • DIX Ethernet used 10 mbps baseband signaling • 802.3u standard was developed to support 100 mbps Ethernet (Fast Ethernet) IEEE 802.5 Standards: • Refers to IBM Token Ring technology • Uses physical star & logical ring thru MAUs with token-passing media access • Can use STP, UTP or Fiber-Optic cables • Data transmission rate is 4 mbps to 16mbps • Mainly used in IBM Mainframes • Uses a single token that travels in only one direction • The network gets slower as the network traffic increases • Data signals are read, amplified and repeated by every device on the network that reduce degradation IEEE 802.12 Standards: • Developed by Hewlett-Packard (HP) • Also known as 100VG, 100VG-AnyLAN, 100BaseVG & AnyLAN (VG = Voice Grade, 100 = 100 mbps) • Combines the concept of Ethernet and Asynchronous Transfer Mode (ATM) • Uses Demand Priority method of Media Access in physical star topology • Uses intelligent hubs to allocate more bandwidth to high-priority data frames

• Both Ethernet & Token Ring frame types are supported by 100VG-AnyLAN Ethernet: • Network architecture (LAN Technology) that shares transmission media • Operates at Physical & Data Link layers of the OSI Model • Signal type – baseband • Physical Topology – Bus or Star • Logical Topology – Bus • Media Access – CSMA/CD • Cables used – Coax, TP or Fiber-Optic • Bandwidth – 10 mbps • Maximum frame size – 1518 bytes IEEE Ethernet Standards: A special 3-part naming conventions – • Data transmission rate = mbps • Signal mode = baseband / broadband • Maximum media distance (meters / yards / feet) or media type Comparison Chart for Ethernet Types: Ethernet Types 10BaseT 10Base2 Physical Topology Star Bus Media Access Method CSMA/CD CSMA/CD Cable type Cat 3 - 5 50-Ohms, UTP RG-58 Coax Transmission speed 10 mbps 10 mbps Max. segment length 100 meters 185 meters Min. segment length 0.5 meters 0.5 meters Max. no. of segments 1024 5 Max. segments with nodes 1024 3 Max. nodes per segment 2 30 Max. nodes per network 1024 90 Max. hubs/repeaters between nodes 4 4 Max. drop cable distance NA 50 meters Comparison Chart for Ethernet Types: (Contd….) Ethernet Types 100BaseTX 100BaseT 4 Physical Topology Star Star Media Access Method CSMA/CD CSMA/CD Cable type Cat 5 UTP Cat 3 – 5 UTP Transmission speed 100 mbps 100 mbps Max. segment length 100 meters 100 meters

Min. segment length Max. no. of segments Max. segments with nodes Max. nodes per segment Max. nodes per network Max. hubs/repeaters between nodes

10Base5 Bus CSMA/CD 50-Ohms, Thick Coax 10 mbps 500 meters 2.5 meters 5

10BaseFL Star CSMA/CD Fiber Optic

100BaseT Star CSMA/CD UTP

10 mbps 2000meters NA 1024

100 mbps 100 meters 2.5 meters 1024

3

1024

1024

100

2

2

300

1024

1024

4

4

4

50 meters

NA

NA

100BaseFX

100VG-AnyLAN

Star CSMA/CD Fiber Optic

2.5 meters 1024

2.5 meters 1024

2.5 meters 1024

Star Demand Priority Fiber Optic, Cat 3-5 UTP, STP 100 mbps Cat 3 UTP: 100 meters Cat 5 UTP: 150 meters STP: 100 meters Fiber Optic: 2000 meters 2.5 meters 1024

1024

1024

1024

1024

2

2

2

2

1024

1024

1024

1024

4

4

4

4

100 mbps 2000 meters

Comparison Chart for Ethernet Types: (Contd….) Ethernet Types Token Ring Physical Topology Star Media Access Method Token Passing Cable type STP, UTP, Fiber Optic Transmission speed 4 or 16 mbps Max. segment length STP = 45 meters UTP = 101 meters Min. segment length 2.5 meters Max. no. of segments 33 Max. segments with nodes 33 Max. nodes per segment Variable Max. nodes per STP = 260 network UTP = 72 Max. hubs/repeaters between nodes 4

FDDI Ring Token passing Fiber Optic 100 mbps 100,000 meters 2.5 meters No Limit NA NA 500 4

TOKEN RING (IEEE 802.5): FIBER DISTRIBUTED DATA IINTERFACE (FDDI – IEEE 802.8): VOICE GRADE (VG) ANYLAN (IEEE 802.12): ASYNCHRONOUS TRANSFER MODE (ATM): APPLETALK: LOCALTALK: ETHERTALK: TOKENTALK:

•

ARCNET: • Full form – Attached Resource Computer Network • Developed by Datapoint Corporation in 1970s • Provides automatic reconfiguration to overcome various faults in a small network • Economical, simple and robust technology • Can support 2047 nodes per network • Data transmission speed – 20 to 100 Mbps • Uses token passing method of media access over logical bus topology • Physical topology – star or bus • Computers must possess token before transmission of data • Data packet contains up to 508 bytes of data • ARCnet Addressing: i. Every ARCnet NIC has 8-bit number as a unique address ii. Range of 1 – 255 is used for node addressing, 0 is used for broadcasts iii. Token passing starts with the machine having the lowest node address ARCnet with Coaxial Configuration: i. Coaxial Cables used – RG-62 (93 Ohm) with BNC connector ii. Maximum number of nodes – 255 iii. Maximum network distance – 6000 m iv. Maximum segment distance – 300 m (using active hub) v. Maximum segment distance – 30 m (using passive hub) ARCnet with Twisted Pair Configuration: i. UTP with RJ-11 or RJ-45 Connectors ii. Maximum number of cascaded hubs – 8 iii. Maximum number of nodes – 255 iv. Maximum network length – 6000 m v. Maximum segment length – 120 m

• •

DATA LINK LAYER DEVICES: Switches Bridges Switches: Operate at the Data Link Layer More intelligent than hubs – • Inspect the data packets as they are received • Determine the source and the destination address of that packet Appropriate delivery of the packet – conserve network bandwidth Ethernet networks rely on contention-based technology (more collision with more devices) To avoid Ethernet collision switched Ethernet is used (hubs are replaced by switches) • HUBS – Single Collision Domain, Single Broadcast Domain, Division of Bandwidth • SWITCHES – Multiple Collision Domain, Single Broadcast Domain, Equal Bandwidth Access

Function of a Layer 2 Switch: • Address Learning • Forward/Filter Decisions Address Learning: • Application Specific Integrated Circuits (ASIC) Chip is used to build and maintain filter table (MAC Table) • When a switch is powered on MAC Table is empty • When a device transmits, the connected interface (port) receives the frame

• •

• • •

The frame’s source address is placed in the MAC Table against the corresponding interface The frame is then flooded out to every port except the source port as the destination port is unknown If a device responds against this broadcasted (flooded) frame, the ASIC chip promptly records the source address of the responded frame along with the corresponding port number into the MAC Table The two devices can now make a point-to-point connection and none other attached devices will receive the frame If the two point-to-point devices don’t communicate again within a certain amount of time, the switch will flush their entries from the MAC database

Forward/Filter Decisions: • When a frame arrives at switch port, the destination MAC Address is compared to the MAC database • If the MAC Address matches, the frame is only sent out to the correct exit interface • If the destination MAC Address is not matched, the frame is flooded out (broadcasted) to all active interfaces except the source interface

LAYER – 3: THE NETWORK LAYER FUNCTIONS: • Logical addressing • Switching • Routing • Network flow control (congestion control) • Network sequencing The Network layer transports traffic between devices that are not locally attached by – • Managing device addressing • Tracking device locations on the network • Determining the best path to move data DEVICES: • Routers • Brouters • Layer 3 Switches LOGICAL ADDRESSING (IP ADDRESSING) Most networks communicate using protocols that must have their own addressing schemes – • TCP/IP (Transmission Control Protocol / Internet Protocol) • IPX (Internetwork Packet Exchange)

1.

IP Terminology Bit – a binary digit, either 1 or 0 2. Byte – a group of 8 bits arranged in a sequence. A byte may also consists of 7 bits depending on whether parity is used 3. Octet – an arrangement of 8 bits. Byte & Octet are completely interchangeable terms IP Address  This is a numeric identifier assigned to each machine on an IP network  It designates the specific location of a device on the network  It is a software address  An IP Address allows a host on one network to communicate with a host on a different network, regardless of the type of LANs  Currently we are using version 4 of IP addressing that consists of 32 bits of information  These 32 bits are divided into 4 sections of 8 bits (called octets), each separated by a period called a DOT  An IP address can be depicted into one of the 3 methods: • Dotted-decimal: 172.16.30.56 • Binary: 10101100.00010000.00011110.00111000 • Hexadecimal: AC.10.1E.38 Note: IP address in Hex form is stored in Windows Registry  The IP addressing scheme is structured in a three-level hierarchical order: (Class A & B) • Network part –> network address • Subnet part –> subnet address • Host part –> host address  In some cases, two-level hierarchical structure is also used: (Class C) • Network part • Host part Network Address (or Network Number) • It is the numeric identifier that uniquely identifies each network • This is the designation used in routing to send packets to a remote network • Every machine on a same network shares that network address as part of its IP address Node Address (or Host Address) • It is assigned to uniquely identify each machine on a network Broadcast Address

•

It is the numeric identifier used to send information to all nodes over a network, or to all nodes of all networks

Example IP address Network address Host address Broadcast address

: 10.16.2.4 : 10.0.0.0 : 0.16.2.4 : 10.255.255.255

172.16.30.56 172.16.0.0 0.0.30.56 172.16.255.255

192.168.10.20 192.168.10.0 0.0.0.20 192.168.10.255

Classification of IP Address Based on the rules defined by InterNIC IP addresses can be classified into following 5 network classes: Class A: Class B: Class C: Class D: Class E:

8 bits 8 bits 8 bits 8 bits N/W Host Host Host N/W N/W Host Host N/W N/W N/W Host Reserved for Multicasting Reserved for scientific R&D

To ensure efficient routing, internet designers defined a mandate for the leading-bits section (leftmost octet or 1st octet) of the IP address for each different network class InterNIC defines that the left-most bits (also called the Most Significant bits) of the 1st octet must be in the following order: Class A = 0 Class B = 10 Class C = 110 Class D = 1110 (multicasting) Class E = 11110 (scientific R&D) The decimal value of any octet is based on the number of bits that are turned on or off For example: when all the bits are turned off, we get – 00000000 = 0x27+0x26+0x25+0x24+0x23+0x22+0x21+0x20 = 0+0+0+0+0+0+0+0 = 0 Again, when all bits are turned on, we get – 11111111 = 1x27+1x26+1x25+1x24+1x23+1x22+1x21+1x20 128+64+32+16+8+4+2+1 = 255 But it is very important to note that all bits of the network part and all bits of the host part cannot be turned all 0s or all 1s, as all 0s means nothing and all 1s means the broadcast. Network Address Range: Class A Network part = 8 bits 1st bit = 0 (turned off) All 0s all 1s excluded So, number of network addresses = 27 - 2 = 126 Again, network address range = 00000001 to 01111111 = 1 to 127 But, 127 is reserved for loop-back testing Hence, Class A network address ranges from 1 to 126 Host Address Range: Class A Host part = 24 bits All 0s all 1s excluded So, number of host addresses = 224 - 2 = 16777216 – 2 = 16777214 Again, host address range = 00000000.00000000.00000001 to 11111111.11111111.11111110 = 0.0.1 to 255.255.254 Network Address Range: Class B Network part = 16 bits 1st bit = 1 (turned on) & 2nd bit = 0 (turned off) All 0s all 1s excluded So, number of network addresses = 214 = 16384 Again, network address range = 10000000.00000000 to 10111111.11111111 = 128.0 to 191.255 Hence, Class B network address ranges from 128 to 191

Host Address Range: Class B Host part = 16 bits All 0s all 1s excluded So, number of host addresses = 216 - 2 = 65536 – 2 = 65534 Again, host address range = 00000000.00000001 to 11111111.11111110 = 0.1 to 255.254 Network Address Range: Class C Network part = 24 bits 1st bit = 1 (turned on), 2nd bit = 1 (turned on) & 3rd bit = 0 (turned off) All 0s all 1s excluded So, number of network addresses = 221 = 2097152 Again, network address range = 11000000.00000000.00000000 to 11011111.11111111.11111111 = 192.0.0 to 223.255.255 Hence, Class C network address ranges from 192 to 223 Host Address Range: Class C Host part = 8 bits All 0s all 1s excluded So, number of host addresses = 28 - 2 = 256 – 2 = 254 Again, host address range = 00000001 to 11111110 = 1 to 254 SUBNETTING IP ADDRESSES Subnetting – logical division of physical IP addresses; a method to break down a larger network into a bunch of smaller ones Benefits: • Reduced network traffic • Optimized network performance • Simplified network management • Efficient connection of large geographical distances Steps for Creating Subnets: • Determining the Class of Network Address from the given IP Address • Determining the no. of Subnets required • Determining the no. of Host bits to be borrowed for Subnetting • Determining the no. of Hosts per Subnet • Determining the Subnet Mask • Determining the valid Subnets • Determining the Host Range and Broadcast Addresses for each Subnet Determining Host (Borrowed) Bits: • Some bits can be borrowed from the host part of the Network ID to create Subnet bits • Borrowed bits are called MASKED BITS (or SUBNET BITS) • Balance Host bits are called UNMASKED BITS Network Bits + Host Bits = 32 Network Bits + (Masked Bits + Unmasked Bits) = 32 If,

Network Bits Masked Bits Host Bits Subnet Hosts/Subnet

=a =b =c =s =h

We have, a + b + c = 32 Again, 2b – 2 >= s

and

2c – 2 >= h

Determining Valid Subnets: • Valid Subnets = 256 – Subnetted Subnet Mask

MORE IS THE NUMBER OF SUBNETS, LESS IS THE NUMBER OFHOSTS PER SUBNET AND VICE VERSA Determining Host Range and Broadcast ID: • Host range starts by turning ON Unmasked bits from the right to the left • 1st Host ID of a valid subnet = very next value of the valid subnet • Last Host ID = subnet ID + no. of Hosts per subnet • Broadcast ID = very next value of the last Host ID SWITCHING ROUTING

LAYER – 4: THE TRANSPORT LAYER FUNCTIONS: • Provides reliable and unreliable delivery (flow control) • Performs error correction before transmission • Keeps data in the form of segments • Multiplexing connections • Protocols – TCP, UDP, SPX

LAYER – 5: THE SESSION LAYER FUNCTIONS: • Provides end-to-end connection between communicating application programs • Allows two networked resources to hold ongoing communication across a network • Applications on each end of the session are able to exchange data for the duration of the session • This layer establishes, synchronizes, maintains & terminates a connection (session) • Practical functions such as security authentication, connection ID establishment, data transfer, acknowledgements and connection release take place • Provides dialog control – keeps different application data separate • Keeps data in the form of messages • Services such as Remote Procedure calls (RPC) & Network File System (NFS) and API such as NetBIOS are typically used on the Session Layer In a nutshell, functions of Session Layer can be classified under 3 broad heads – 1. Connection Establishment 2. Data Transfer 3. Connection Release Connection Establishment: i. The sender tries to communicate to the receiver ii. The receiver asks for connection authentication from the sender iii. The sender provides with the user login name and password for authentication iv. Once the sender is recognized, they both agree to communicate v. A connection ID number is established using the Transport Layer Service Address vi. They then agree upon the type of service required and the duration of service vii. Then the entity that is going to begin the conversation is determined viii. Both the end stations coordinate upon message sequencing, acknowledgement and retransmission procedures, represented as Session Layer Error Checking ix. This layer maintains the line of communication and ensures reliable conversations Once the above tasks have been accomplished and the connection has been opened, data transfer begins Data Transfer: i. Session layer messages are controlled and ensures productive conversations ii. Data transfer occurs through a specific Dialogue Control between the two networked entities iii. Dialogues can occur in one of the three ways – • Simplex • Half-Duplex • Full-Duplex iv. During data transfer, the session layer is responsible for ACK or NACK v. This reliability feature ensures that lower-layer protocols keep track of which messages belong to which upper-layer services vi. This layer provides error-checking algorithm that calls for the resumption of interrupted communication vii. Data transfer continues throughout the life of the conversation Simplex Communication: • It is only a one-way communication • Only one device is allowed to transmit, while all other devices can only receive • The transmitting device cannot receive and the receiving devices cannot transmit • Full bandwidth of the channel is utilized for transmitting the signal Half-Duplex Communication: • Each device can transmit and receive, but not at the same time • The transmitting device cannot receive while transmitting and the receiving device cannot transmit while receiving • Full bandwidth of the channel is available to the transmitting device

•

Only one channel is required for both transmission and reception and support bi-directional communication

Full-Duplex Communication: • Every device is allowed to transmit and receive at the same time • Two channels are required – one for transmission and the other for reception Once the communication have ended, the connection must be released Connection Release: • Connection must be released to make room for future conversations • The process can be accomplished by either of the two – a) Planned Connection Release b) Accidental Connection Release Planned Connection Release: • It is accomplished by mutual agreement between the two systems Accidental Connection Release: • It is caused by physical anomaly – an obvious loss of connection • Lower-layer protocols recognize a loss of connection when they receive a NACK • The lower-layer protocols inform the Session Layer that the connection has been lost • The error-checking algorithm at the Session Layer resumes the interrupted dialogue using a new conversation and resolves the accidental connection release Once the networking connection has been established and data transfer begun, focus shifts to the Presentation Layer, where the data is translated into a mutually agreed-upon format

LAYER – 6: THE PRESENTATION LAYER FUNCTIONS: • Provides appropriate data presentation • Provides data encryption, compression and translation services • Keeps data in the form of messages • Services – ASCII, JPEG, MPEG, GIF, TIFF, RTF, BMP, MIDI, etc.

LAYER – 7: THE APPLICATION LAYER FUNCTIONS: • Provides a user interface • Provides file, printer, message, database & application services • Defines how interaction occurs between network services (applications) and the network • Keeps data in the form of messages • Services – telnet, ftp, web browsers, e-mails, word processing, etc.

NET WORK CARD DRIVERS

•

Driver: Software used by a hardware device to communicate with the Operating System

•

NIC Driver: Software used by the NIC to communicate with the Operating System. Also called MAC Driver

Sources of Drivers: • Operating System in-built drivers • Model-specific drivers supplied by the NIC manufacturers (support different Operating Systems) • Newer and upgraded version available from the vendors’ web sites

o Note: latest version on drivers provide –  Least number of bugs  Optimized performance  Reduced problems Hardware Driver Installation: Manual Installation Start -> Settings -> Control Panel -> ‘Add New Hardware’ Automatic Installation ‘New Hardware Found’ -> Insert Driver Disk As per of OS Installation Resources Required: • NIC drivers use certain computer resources – • RAM • DMA Channel • I/O Port • IRQ Driver Interfaces: • Open Driver Interface (ODI) • Network Driver Interface Specifications (NDIS) Open Driver Interface (ODI): • Also called Multiple Link Interface Driver (MLID) • Developed by Novell • Bound to only one protocol stack Network Driver Interface Specifications (NDIS): • Developed by Microsoft Corporation • Allows multiple protocol stacks to be bound to multiple NICs Note: ODI & NDIS are incompatible to each other TRANSPORT PROTOCOLS Protocol: • Agreed-upon format of messages at each layer • Rules and conventions for transmitting data between two devices • Determines the type of error checking process • Determines the data compression method • Determines the method of acknowledgement Protocol Stack / Protocol Suite: • Collection of protocols and the order in which they work together • Communicating machines must use a common protocol stack for flawless communication • Most common protocol suites are –  TCP/IP

 IPX/SPX  NetBEUI TRANSMISSION CONTROL PROTOCOL / INTERNET PROTOCOL (TCP / IP): Transmission Control Protocol (TCP) – • Main Transport layer protocol • Provides Service Addresses (Port numbers) • Provides reliable, full-duplex, connection-oriented transport service to upper-layer protocols • Works in conjunction with IP to move packets to move packets through the Internetwork • Assigns connection ID (port number) to each virtual circuit • Provides message fragmentation & reassembly using sequence numbering • Error checking is enhanced through TCP acknowledgement Internet Protocol (IP) – • A connectionless datagram protocol that works at the Network layer • Uses logical network address • Uses packet-switching method • Uses dynamic routing table for route selection that are referenced at each hop • If a connection goes down or become congested, packets could take different route using IP address • Provides error control for connection services Information required for Configuring TCP/IP: • IP Address –  Provides node address  Provides network address Subnet Mask – Determines local or remote network Default Gateway – IP address of the attached router interface Limits the device to communicate only within the local network

 

Domain Name System (DNS) – A distributed database system that works at the Transport layer Provides name-to-address mapping for client applications

Note: DNS servers maintain databases that consist of hierarchical name structures of the variable domains in order to use logical names for device identification. This type of address/name resolution is called service-provider initiated. The largest use of DNS is in the Internet. Name Servers (DNS servers) are used to translate site names to actual network addresses. •

Windows Internet Naming Service (WINS) –  Microsoft NT Network version of DNS  Functions similar to the DNS •

•

Dynamic Host Configuration Protocol (DHCP) –  A protocol that works on a TCP/IP network  DHCP Server maintains a pool of valid IP addresses & other settings  IP addresses are assigned automatically  These IP addresses are allocated to computers to access remote networks  IP addresses are allocated for a predetermined period of time (lease period)  Allows network administrators to supervise and distribute IP addresses from a central point DHCP Operation:  An IP address is requested thru TCP/IP operation in the form of a broadcast  DHCP server receives the request  A new address is assigned for a leased period along with other information  Receipt is acknowledged and configuration being set  During the lease period – o DCHP server will not reallocate the address

o   by –

Returns the same address every time a new address is requested Lease period can be extended by giving subsequent requests IP address may be released before the expiry of the lease period

o Telling the DHCP server that it is no longer required  The released IP address can be used by other client on the network DYNAMIC HOST CONFIGURATION PROTOCOL For the internet to function properly some protocols are used These protocols are defined by a volunteer organization called Internet Engineering Task Force (IETF) A division of IETF called Dynamic Host Configuration Working Group created DHCP DHCP is given the status as an Internet Standard by Internet Activities Board (IAB)