Comp2410 Notes PDF

Preview

Summary

notes...

Description

COMP2410 NOTES CONTENTS

Lecture 1.....................................................................................................................................1 Assessment.............................................................................................................................1 Data Communications............................................................................................................1 Tutorial 1 (Online)......................................................................................................................1 Lecture 3.....................................................................................................................................3

LECTURE 1

ASSESSMENT  



Weekly Quiz (0-5 marks) Assignment (0-25 marks) o From week 7-11 o Individual Final Exam (0-70 marks) o 28 mark (40%) hurdle

DATA COMMUNICATIONS   

Data communications is the movement of computer information from one point to another by means of electrical or optical transmission systems Telecommunications is a broader term for long-distance transmission of voice, video (images and graphics), and data Basic Computer Components o Client is a user device to access network and receive data o Server is a device that stores and transmits data to a client o Circuit is a pathway or connection between client and server

TUTORIAL 1 (ONLINE)

1. How can data communication networks affect businesses? They are the foundations for distributed systems in which information system application are divided among a network of computers. Facilitate more efficient use of computers and improve the day-to-day control of a business by providing faster information flow. 2. Discuss three important applications of data communication networks in business and personal use. Email, Skype, Internet

3. Define the acronyms of LAN, BN, MAN, and WAN and explain how LANs differ from WANs, and BNs? LAN, Local Area Network: group of terminals located in same general area BN, Backbone Network: large central network that connects almost everything on a single site MAN, Metropolitan Area Network: encompasses a city or county area WAN, Wide Area Network: spans cities, states or national boundaries 4. What is a circuit? A pathway or connection between client and server 5. What is a client? User device that accesses a network and receives data 6. What is a server? Device that stores and transmits data 7. Why are network layers important? Layers enable simplicity in development and comprehension of complex networks. 8. Describe the seven layers in the OSI network model and what they do. Application: software used by the network user Presentation: formats the data for presentation to the user by accommodating interfaces on different terminals or computers Session: responsible for initiating, maintaining and terminating the logical session between end users Transport: Deals with the end-to-end issues such as procedures for entering and departing from a network Network: Takes the message generated by the application layer and if necessary, breaks it into several smaller messages. Then address the messages and determines their route through the network and records message accounting information before passing it to the data link layer. Data Link: formats the message to indicate where in starts and ends, decides when to transmit it over the physical media and detects and corrects any errors Physical: physical connection between the ender and receiver, including hardware devices 9. Describe the five layers in the Internet network model and what they do. Application: software used by network user Transport: deals with end-to end issues Network: Data Link: Physical: 10. Explain how a message is transmitted from one computer to another using layers. Application layer is used by the end user. Transport layer obtains the address of the end user, breaks a large data transmission into smaller packets, ensures that all packets

have been received, eliminates duplicate packets. Network layer takes the message generated by the app layer and breaks it into smaller messages if necessary, It then addresses these messages and determines their route, records message accounting info before passing it on. Data link layer formats the message to indicate where is begins and ends, decides whether to transmit over physical media. Physical layer facilitates the transmission. 11. Describe the three stages of standardization. Please describe the formal standardization process, explain how ISO standards are developed, and give examples of national standards bodies mentioned in the chapter. 12. How are Internet standards developed? 13. Describe two important data communications standards-making bodies. How do they differ? 14. What is the purpose of a data communications standard? 15. What are three of the largest inter-exchange carriers (IXCs) in North America? 16. Discuss three trends in communications and networking. 17. Why has the Internet model replaced the OSI model? 18. In the 1980s, when we wrote the first edition of this book, there were many, many more protocols in common use at the data link, network, and transport layers than there are today. Why do you think the number of commonly used protocols at these layers has declined? Do you think this trend will continue? What are the implications for those who design and operate networks? 19. The number of standardized protocols in use at the application layer has significantly increased from the 1980s to today. Why? Do you think this trend will continue? What are the implications for those who design and operate networks? 20. How many bits (not bytes) are there in a 10 page text document? Hint: There are approximately 350 words on a double-spaced page. LECTURE 3

THE APPLICATION LAYER 





Example: clicking on a webpage o Purpose: User’s access to network, software to perform work o Protocols/Standards  HTTP, SMTP, DNS, FTP, DHCP, IMAP, POP, SSL o Protocol Data Unit (PDU)  Packet or Data Application Architecture o The way the function of the application layer are spread out across the client and server Functions of the application layer o Presentation logic  Presentation of information to the user and acceptation of the user commands







o Application/business logic  Everything the application does o Data access logic  Processing required to access data, often SQL queries o Data storage  Storage and retrieval of small files and/or large databases Host-based architectures (1960s) a.k.a. server based o Simple o Single point of control o Host can become a bottleneck o Upgrades typically expensive Client based architectures (1980s) o Simple o Hardware and application less expensive o Data must travel back and forth between server and client Client-server architectures (2010s) o More efficient because of distributed processing o Allows hardware/software to be used together o Less bandwidth required









o Configuring hardware/software from different vendors to work together can be hard Middleware o Provides a standard way of translating between software from different vendors o Manages message transfers o Insulates network changes from the clients N-Tier architecture with four computers o Client (presentation logic) o Web server (application logic) o Application server (application logic o Database server (data access logic and data storage) Thin vs thick client architecture o Thick – application logic on client o Thin – application logic on server Three models of cloud computing o Software as a service (SaaS)  All application components and associated hardware/software are outsourced  Gmail, Salesforce o Platform as a service (PaaS)



 Application logic and data are managed internally  Amazon Elastic Cloud Compute. Microsoft Azure o Infrastructure as a service (IaaS)  All hardware is outsourced  Amazon Web Services, Microsoft Azure Choosing an architecture o Infrastructure  Cost of servers, clients and circuits  Reliability o Development costs  Software; cheaper on host-based architectures o Scalability  

Ability to increase or decrease in computing capacity as network demand changes Easier in client-server architectures

LECTURE 4

NETWORK AND TRANSPORT LAYERS 

Transmission Control Protocol (TCP)/Internet Protocol (IP) o Developed as a single protocol in 1974 o Divided later into each o Most common protocols of the internet







o Allows reasonably efficient and error-free transmission o Compatible with a variety of data link protocols TCP o Most common transport layer protocol o A connection-oriented communications protocol o 192-bit (24 byte) header of control information User Datagram Protocol (UDP) o Second type of transport layer protocol o Usually used for single packet transmission o Suitable for real-time applications IP o Network layer protocol o IPv4/IPv6

LECTURE 5

LOCAL AREA NETWORK 

Why use a LAN?



o Resource (e.g., printers) sharing o Information sharing  Improved decision making  May reduce data duplication and inconsistency o Software sharing  Some software (e.g., Matlab) can be purchased on a per-seat basis and resides on server  Reduces costs, simplifies maintenance and upgrades o Easier device management (e.g., software updates and configuration) Components o Clients o Servers o Network Interface Cards









o Network Circuits o Hubs/switches/access points o Network Operating systems Clients and Servers o Computer hardware and/or software that access services made available by servers o Computer programs and/or devices that deliver information or provide services to clients Network Interface Card o Provides network connection to the physical layer of a computer. o Also called network interface controller, network adapter or LAN adapter. o Operate at physical and data link layers. o Commonly built into motherboards. o Can be internal as well as external. Network Circuits o Each computer is connected to other computers using the circuit o Use unshielded twisted pair (UTP) cable, shielded twisted pair (STP) cable or fibre optic cable o Use radio transmissions to send data between the NIC and the access point (AP) Hub/switch/router o Hubs and switches are used to create networks where routers are used to connect networks o Hubs and Switches  Work as junction boxes  Act as repeaters  Used to send message from one port to other ports











 Receiver decides whether to accept the message or not  Knows the specific addresses where to send the message  More efficient than a hub and provide faster network speed  Port is a connection point where a cable to plugged in  Each port has a unique number Access Point o A radio transceiver that plays the same role as a hub or switch o Creates a wireless local area network (WLAN) o Wired via an Ethernet cable o Projects a Wi-Fi signal to a designated area Network operating systems o An OS that manages network resources and connects computers and devices into a LAN o It may be included with the client operating system itself (mac, windows, linux)  Or it may require an existing OS in order to function (LAN servicer requires OS/2; LANtastic requires DOS) o NOS server software  Performs tasks associated with data link, network, and application layers  Enables various servers, e.g. file server, print server, database server, etc. o NOS Client Software  Performs tasks associated with data link and network layers  Provides directory services for getting information about available resources, such as Microsoft’s Active Directory Service (ADS) o Network profiles  Specifies available resources for use  Need to include on the network profile assignment list for being available to use Network topology o Logical topology: How the network is conceptually connected o Physical topology: How the network is physically installed o Pros and Cons of Wired vs wireless slide 20 Wired Ethernet o Layer 2 protocol, i.e. operates at data link layer  Physical layer must meet its requirements  IEEE 802.3 standards are used nearly by all LANs Hub based ethernet o Bus topology: passes data to all computers through the central cable (bus) o Star topology: all cables are connected to the central hub

o Main feature: share a same multipoint circuit, called collision domain



Switch based ethernet o Star topology - Logical and physical both o Uses a forwarding table - only C will receive the frames o Layer 2 switch - use data link layer protocol



Switching modes o Cut-through switching: low latency and worse error control o Store-and-forward switching: high latency and better error control o Fragment-free switching: compromise between the two



Media Access Control o It is a way to prevent a collision or to recover from a collision o Carrier Sense Multiple Access with Collision Detection (CSMA/CD)  Listen while transmitting  Wait until the circuit is free and then transmit Wireless Ethernet o Using radio frequencies to transmit signals through the air o Wi-Fi has many benefits  Network connection where cabling is impossible or undesirable  Allows device and user mobility  Potentially more economical than wired networks o IEEE 802.11 standards Network Topology o Physical star o Logical bus Media access control







o Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) - listen before transmitting o Scanning  Active Scanning: NIC transmits probe frame to be connected  Passive Scanning: NIC listens for beacon frame to be connected o Methods  distributed coordination function (DCF)/ physical carrier sense method  stop-and-wait ARQ  point coordination function (PCF)/ virtual carrier sense method  overcome hidden node problems  use controlled-access technique

Two extra addresses are needed (transmitter and receiver addresses) as NICs communicate via AP  Segmentation is done in data link layer if needed Types of Wireless Ethernet 





o Security o Wired Equivalent Privacy (WEP)  all data is encrypted before sending  easy to break o MAC Address Filtering  AP only passes data to the listed MAC address  easy to break o Wi-Fi Protected Access (WPA)  every frame is encrypted









 hard to break o 802.11i (also called WPA2)  master key is used for each session  most secure LAN design – Best practise o Primary network  Wired LAN  Usually for desktops o Overlay network  Wi-Fi  Usually for moving devices Wireless LAN – Practical use o Primary network  Password Protected  Used by regular users  Less traffic o Guest network  Password protected  Authorised users only (e.g. airport, market)  Huge traffic Designing the Data Centre o Load Balancer  add/remove server(s) without affecting users o Server Virtualisation  separate logical servers on the same physical computer o Storage Area Network (SAN)  high-speed storage device o Network-Attached Storage (NAS)  designed to response for files and data on SAN LAN Performance o Performance Measurement  Throughput: the total amount of user data transmitted in a given time period  Response time: how long a computer takes to get a response from the destination o Bottleneck @Network server, if the server utilization is high (e.g., 80–100%) during poor performance  @Network circuit, if the server utilization is low during poor performance o Improving 

 





Software:  upgrading NOS Hardware:  upgrading server(s)/ sharing works among servers  increasing amount of memory in the server(s)  increasing number of hard disks in the server(s)/ using RAID* Circuit Capacity  Buy a faster circuit  Add a circuit  Add APs on different channels  Segment the network Reducing Network Demand  Move files to client computers  Encourage off-peak usage  Switched from wired to wireless, or vice versa

LECTURE 6 – BACKBONE NETOWKRS









Network cables as a backbone component o Often fibre cables for higher data rates (than for LANs) o If buildings > 100 m apart, twisted-pair cables cannot reach Switches as a BN component o Operate on the Data Link Layer to connect devices within a network (same subnets):  Layer-2 switches are “transparent” devices that only read and forward messages WITHOUT changing them  VLAN switches or layer-3 switches are devices, which combine the features of Layer-2 switches and routers Routers as a BN components o Operate on the Network Layer using layer-3 addresses to connect different subnets  read IP addresses  determine the best route which requires more processing than switches can offer  can respond to messages (e.g., ARP)  must strip off frames from the Data Link Layer before passing messages up to the Network Layer Major BN Architectures o Switched BN o Routed BN o VLAN BN o Most commonly a combination of the three



Terminology o Rack-mounted equipment: The rack of equipment where all network devices are physically located o Chassis switch: It enables users to plug modules directly into the switch o Main distribution facility (MDF) or central distribution facility (CDF): The room containing the rack of equipment o Patch cables: The very short cables that connect devices in the rack o Module: A certain type of network device, such as 16-port 100Base-T switch, router, or 24-port1000Base-T switch, etc.

LECTURE 7

VIRTUAL LANS AND METRO AREA NETWORKS 







Packed switched networks o Data is "packetized" into groups of bytes/octets o Ethernet (layer 1,2) called "frames" o TCP/UDP/IP (layer 3,4 etc.) called "segments" o Asynchronous Transfer Mode (ATM) (all layers) called "cells" o Each packet routed through network infrastructure separately o Advantages o Disadvantages IEEE802 Protocols o Ethernet, WiFi, Bluetooth etc. use IEEE802.x standards for packetization o packets called frames o variable length, up to 1500 bytes data + headers ("jumbo" frames also allowed) o on Ethernet, several frame "types":  IPv4 and 6 typically uses "Ethernet II" (DIX) frames (14-byte header, 4 byte CRC)  other protocols use 802.2 LLC and 802.2 SNAP encapsulation o Advantages Ethernet 2 o EtherType (or "length") field is >= 1536 (greater than max. data allowed)  0x0800 (2048) for IPv4  0x0806 for Address Resolution Protocol (ARP)  0x86DD for IPv6 o Ethernet II formally recognised in IEEE802.3x-1997 and later standards Broadcast Domain o a group of networked nodes that will receive frames sent to layer 2 broadcast address o (or that can participate in a multicast group at layer 2)

o traffic between nodes in different broadcast domains necessarily must be handled at layer 3 o sometimes called Local Area Networks, subnets, layer 2 etc.



o generally all nodes in a broadcast domain will have the same IPv4 network part, but different host parts Ethernet Switches o almost all Ethernet these days is point-to-point with one or more "switches" in the centre of a star o home networks usually have a router/wifi access point with a small Ethernet switch built in o switches can be unmanaged or managed o a managed switch allows configuration and monitoring etc. o managed switches usually allow the ports to be grouped into different



broadcast domains o such "virtual" networks are called VLANs o how to join two or more managed switches and extend those broadcast domains? IEEE802.1Q tags o defines a new EtherType: 0x8100 o allows frames between com...