Presentation is loading. Please wait.

Presentation is loading. Please wait.

Computer Network Basics Components of Any Computer Processor (active) Computer Control (“brain”) Datapath (“brawn”) Memory (passive) (where programs,

Similar presentations


Presentation on theme: "Computer Network Basics Components of Any Computer Processor (active) Computer Control (“brain”) Datapath (“brawn”) Memory (passive) (where programs,"— Presentation transcript:

1

2 Computer Network Basics

3 Components of Any Computer Processor (active) Computer Control (“brain”) Datapath (“brawn”) Memory (passive) (where programs, data live when running) Devices Input Output Keyboard, Mouse Display, Printer Disk, Network

4 Communication Devices r Synchronous communication uses a clock signal separate from the data signal- communication can only happen during the ‘tick’ of the timing cycle r Asynchronous communication does not use a clock signal- rather, it employs a start and stop bit to begin and end the irregular transmission of data

5 Connecting to Networks (and Other I/O) r Bus - shared medium of communication that can connect to many devices r Hierarchy of Buses in a PC

6 Operating Systems

7 Operating Systems Developed for Portable Devices

8 A Closer Look at Network Structure r network edge: applications and hosts r network core: m routers m network of networks

9 General Architecture of Computer Networks

10 The Network Core r mesh of interconnected routers r the fundamental question: how is data transferred through net? m circuit switching: dedicated circuit per call: telephone net m packet-switching: data sent thru net in discrete “chunks”

11 Connection of Networks

12 Network Topology a) bus, b) star, c) ring, d) tree structure a)b)c)d)

13 Classification of the networks according to the connection establishing r Line switched network r Packet switched network r Radiating/data disseminating systems r Point-to-point connected networks

14 Wired Media r Telephone line r Thin Coax r Thick Coax r Unshielded Twisted Pair (UTP) r Shielded Twisted Pair (STP) r Fibre

15 (Data) Reliability r A network service is (data) reliable if the sender application can rely on the error-free and ordered delivery of the data to the destination r In the Internet the reliability can obtained mainly by acknowledgements and retransmission r In such a way the losses in the underlying layers can be retrieved

16 Flow-control and Congestion Prevention r Flow-control: to protect the receiver against the overload m I.e.: the sender (source) sends more data than the receiver can process m it is mainly necessary in link and transport level r Congestion prevention: to prevent the intermediate nodes against the overload m it is mainly necessary in network level

17 Overload and Congestion r Overload: Too many packets occur in a subnetwork in the same time, which prevent each other and in such a way the throughput decreases r Congestion: the queues in the routers are too long, the buffers are full. m As a consequence some packages are dropped if the buffers of the routers are overloaded r In extreme case: grid-lock, lock-up

18 Deadlock r Deadlock: the most serious situation of the congestion, the routers wait for each other r Direct store and forward deadlock: the buffers of two neighbouring routers are full with the packets to be sent to the other router r Indirect store and forward deadlock: the deadlock occurred not between two neighbouring routers but in a subnetwork, where any of the routers has not free buffer space for accepting packets

19 r Network: physical connection that allows two computers to communicate r Packet: unit of transfer, bits carried over the network m Network carries packets from on CPU to another m Destination gets interrupt when packet arrives r Protocol: agreement between two parties as to how information is to be transmitted r Broadcast Network: Shared Communication Medium r Delivery: How does a receiver know who packet is for? m Put header on front of packet: [ Destination | Packet ] m Everyone gets packet, discards if not the target r Arbitration: Act of negotiating use of shared medium r Point-to-point network: a network in which every physical wire is connected to only two computers r Switch: a bridge that transforms a shared-bus (broadcast) configuration into a point-to-point network r Router: a device that acts as a junction between two networks to transfer data packets among them Networking Definitions

20 The Need for a Protocol Architecture r Procedures to exchange data between devices can be complex r High degree of cooperation required between communicating systems m destination addressing, path m readiness to receive m file formats, structure of data m how commands are sent/received and acknowledged m etc.

21 Layered Protocol Architecture r Modules arranged in a vertical stack r Each layer in stack: m Performs related functions m Relies on lower layer for more primitive functions m Provides services to next higher layer m Communicates with corresponding peer layer of neighboring system using a protocol

22 Network Layering r Layering: building complex services from simpler ones m Each layer provides services needed by higher layers by utilizing services provided by lower layers r The physical/link layer is pretty limited m Packets are of limited size (called the “Maximum Transfer Unit or MTU: often bytes in size) m Routing is limited to within a physical link (wire) or perhaps through a switch r Our goal in the following is to show how to construct a secure, ordered, message service routed to anywhere: Physical Reality: PacketsAbstraction: Messages Limited SizeArbitrary Size Unordered (sometimes)Ordered UnreliableReliable Machine-to-machineProcess-to-process Only on local area netRouted anywhere AsynchronousSynchronous InsecureSecure

23 Key Features of a Protocol r Set of rules or conventions to exchange blocks of formatted data r Syntax: data format r Semantics: control information (coordination, error handling) r Timing: speed matching, sequencing r Actions: what happens when an event occurs

24 Operation of Protocols

25 The OSI Model r Physical Layer r (Data) Link Layer r Network Layer r Transport Layer r Session Layer r Presentation Layer r Application Layer

26 Physical Layer r Transmission of energy onto the medium m Collection of energy from the medium m This layer is concerned with the physical transmission of raw bits m This bits are transmitted through mechanical, electrical, and procedural interfaces which include interface card standard modem standards certain portions of the ISDN and LAN MAN standards

27 (Data) Link Layer r Transmission of frames over one link or network r Often subdivided into the MAC and LLC r It receives bits from the physical layer, converting bits to frames m frame boundaries r Using protocols (e.g. HDLC), this layer corrects errors that might have occurred during transmission across a link r In addition this layer provides an “error-free” transmission channel to the next layer known as the network layer: error control m ARQ m duplicates r Flow control

28 r The previous two layers were concerned with getting error-free data across a link r The network layer establishes connections between nodes, routes data packets through the network, and accounts for them r End-to-end transmission of packets (possibly over multiple links) r Controls the operation of the subnet r Routing m static m dynamic r Congestion control m At this stage, there may be congestion due to many packets waiting to be routed m Some packets may be lost during congestion Network Layer I

29 Network Layer II r Accounting m packets m bytes m etc. r Internetworking m This layer is also concerned with internetworking where there is ‘talking’ between technologies, such as the traditional Internet connected to ATM m segmentation m addressing m sequencing m accounting r Broadcast subnets: thin network layer

30 Transport Layer I r This layer presumes the ability to pass through a network and provides additional services to end-users, such as and-to-and packet reliability r End-to-end delivery of a complete message (end-to-end communication path, usually reliable) r Isolation from “hardware” r Multiplexing/demultiplexing r Divide message into packets r Reassemble (possibly out of order packets) into the original message of the distant end

31 Transport Layer II r End-to-end flow control r Acknowledgments r Types of service m error-free, point-to-point, in sequence, flow controlled m no correctness guarantees m no sequencing r Establishing/terminating connections m naming/addressing m intra-host addressing (process, ports)

32 r This layer enables users to establish sessions across a network between machines r In addition, it offers session management services r Set up and management of end-to-end conversation r Establish and terminate sessions m superset of connections r Assignment of logical ports r Dialogue control r Token management m for critical operations r Synchronization m checkpoints/restarts Session Layer

33 Presentation Layer r This layer is concerned with the syntax and semantics of messages, code conversions between machines, and other data conversion services r Some of these services are data compression and data encryption r Interface between lower layers and application r Formatting r Syntax & semantics of messages r Data encoding (e.g.: ASCII to EBCDIC) r Compression r Encryption/Decryption r Authentication

34 Application Layer r This layer provides support for the user's network applications r Some application layer services have been standardized, e.g.: m File Transfer and Management (FTAM) m Message Handling Services for electronic mail (X.400) m Directory Services (X.500) m Electronic Data Interchange (EDI) r Program you’re running,applications m file transfer, access & management m m virtual terminals m WWW

35 The OSI Protocol Stack

36 Operation of the model

37 Names of the Nodes, Connections and Data Units

38 Communication among the layers m Connection oriented network service (virtual circuits, eg. ATM) Reliable transport service Unreliable transport service m Connectionless network service (datagram service, eg. IP) Reliable transport service (eg. TCP) Unreliable transport service (eg. UDP)

39 Network Tools r Repeater: connects network segments logically to one network r Hub: multiport repeater r Bridge: datalink level connection of two networks r Switch: multiport bridge r Router: connects networks that are compatible in transport level m subnetworks are connected to the interfaces of the repeater r Gateway (proxy server): router between two individual network. The “Way Out”

40 Physical Layer Devices r Repeater r Hub m “dumb” m level-1 hub m multi-port repeater

41 Data Link Layer Devices r Bridge m Cascaded vs. Backbone m Single m Multiple r Switch (switched hub)

42 Routers r Provide link between networks r Accommodate network differences: m Addressing schemes m Maximum packet sizes m Hardware and software interfaces m Network reliability m Congestion/Traffic Management

43 Devices of the Network Connection

44 Architectural Implementation of the LANs m Ethernet (IEEE 802.3) m FDDI m Gigabit Ethernet m Token Bus (IEEE 802.4) m Token Ring (IEEE 802.5)

45 Characteristics of High-Speed LANs Fast EthernetGigabit EthernetFibre ChannelWireless LAN Data Rate100 Mbps1 Gbps, 10 Gbps 100 Mbps – 3.2 Gbps 1 Mbps – 2 Gbps Transmission Mode UTP,STP, Optical Fiber UTP, shielded cable, optical fiber Optical fiber, coaxial cable, STP 2.4 GHz, 5 GHz Microwave Access MethodCSMA/CD SwitchedCSMA/CA Polling Supporting Standard IEEE Fibre Channel Association IEEE

46 Wide Area Network Connections r Solutions for connecting LANs to the Internet  Ethernet (ring or star topology) m Managed Leased Line Network (MLLN) m ATM (Asynchronous Transfer Mode) m Switched line m ISDN line

47 Soft and Hard States r State: the data collection, which are necessary for keeping the connection between two protocol entities r Hard state m If the connection is established once, it is never timed out, even if it is not in usage m To cancel the connection one of the participants of the connection must explicitly close it m The history of the state is stored r Soft state m To keep the connection the participants must send occasionally keep-alive messages, since without keep-alive message the state information is timed out after a certain period m The state is called as “soft” since in the ordinary operation the state can change easily m The history of the state is not stored

48 Packet switching versus circuit switching r Great for bursty data m resource sharing m no call setup (less start-up delay) r However… m Packets can experience delays, so not for “real-time” applications m excessive congestion leads to packet delay and loss protocols (like TCP) are needed for reliable data transfer, and congestion control Is packet switching best in every case?

49 Performance Considerations r Before continue, need some performance metrics m Overhead: CPU time to put packet on wire m Throughput: Maximum number of bytes per second Depends on “wire speed”, but also limited by slowest router (routing delay) or by congestion at routers m Latency: time until first bit of packet arrives at receiver Raw transfer time + overhead at each routing hop r Contributions to Latency m Wire latency: depends on speed of light on wire about 1–1.5 ns/foot m Router latency: depends on internals of router Could be < 1 ms (for a good router) Router L R1 L R2 L W1 L W2 L w3

50 Delay in packet-switched networks packets experience delay on end-to-end path r four sources of delay at each hop r Nodal processing: m check bit errors m determine output link r Queueing: m time waiting at output link for transmission m depends on congestion level of router A B propagation transmission nodal processing queueing

51 Delay in packet-switched networks Transmission delay: r R=link bandwidth (bps) r L=packet length (bits) r time to send bits into link = L/R Propagation delay: r d = length of physical link r s = propagation speed in medium (~2x10 8 m/sec) r propagation delay = d/s A B propagation transmission nodal processing queueing

52 Queueing delay r R=link bandwidth (bps) r L=packet length (bits) r a=average packet arrival rate traffic intensity = La/R r La/R ~ 0: average queueing delay small r La/R -> 1: delays become large r La/R > 1: more “work” arriving than can be serviced, average delay infinite!

53 Internet protocol stack r Application: supporting network applications m ftp, smtp, http r Transport: host-host data transfer m tcp, udp r Network: routing of datagrams from source to destination m ip, routing protocols r Network access: data transfer between neighboring network elements m ppp, ethernet r Physical: bits “on the wire”

54 Layering: logical communication application transport network link physical application transport network link physical application transport network link physical application transport network link physical network link physical data E.g.: transport r take data from app r add addressing, reliability check info to form “datagram” r send datagram to peer r wait for peer to ack receipt r analogy: post office data transport ack

55 Layering: physical communication application transport network link physical application transport network link physical application transport network link physical application transport network link physical network link physical data

56 Protocol layering and data Each layer takes data from above r adds header information to create new data unit r passes new data unit to layer below application transport network link physical application transport network link physical source destination M M M M H t H t H n H t H n H l M M M M H t H t H n H t H n H l message segment datagram frame

57 IP over ATM r ATM Adaptation Layer (AAL): interface to upper layers m end-system m segmentation/rea ssembly r ATM Layer: cell switching r Physical AAL5 ATM physical AAL5 ATM physical AAL5 ATM physical AAL5 ATM physical ATM physical application TCP/UDP IP application TCP/UDP IP application TCP/UDP IP application TCP/UDP IP

58 Physical Data Link Network Transport Session Presentation Application Network Access IP TCP UDP Application Sockets The Internet Protocol Stack

59 Network Protocols r Protocol: Agreement between two parties as to how information is to be transmitted m Example: system calls are the protocol between the operating system and application m Networking examples: many levels Physical level: mechanical and electrical network (e.g. how are 0 and 1 represented) Link level: packet formats/error control (for instance, the CSMA/CD protocol) Network level: network routing, addressing Transport Level: reliable message delivery r Protocols on today’s Internet: Ethernet ATM Packet radio IP UDPTCP RPC NFS WWW ssh Physical/Link Network Transport

60 Building a messaging service r Process to process communication m Basic routing gets packets from machine  machine m What we really want is routing from process  process Example: ssh, , ftp, web browsing m Several IP protocols include notion of a “port”, which is a 16-bit identifiers used in addition to IP addresses A communication channel (connection) defined by 5 items: [source address, source port, dest address, dest port, protocol] r UDP: The User Datagram Protocol m UDP layered on top of basic IP (IP Protocol 17) Unreliable, unordered, user-to-user communication UDP Data 16-bit UDP length16-bit UDP checksum 16-bit source port 16-bit destination port IP Header (20 bytes)

61 Building a messaging service (con’t) r UDP: The Unreliable Datagram Protocol m Datagram: an unreliable, unordered, packet sent from source user  dest user (Call it UDP/IP) m Important aspect: low overhead! Often used for high-bandwidth video streams Many uses of UDP considered “anti-social” – none of the “well- behaved” aspects of (say) TCP/IP r But we need ordered messages m Create ordered messages on top of unordered ones IP can reorder packets! P 0,P 1 might arrive as P 1,P 0 m How to fix this? Assign sequence numbers to packets 0,1,2,3,4….. If packets arrive out of order, reorder before delivering to user application For instance, hold onto #3 until #2 arrives, etc. m Sequence numbers are specific to particular connection

62 Message TCP/IP packet, Ethernet frame r Application sends message TCP data TCP Header IP Header IP Data EH Ethernet Hdr r TCP breaks into 64KB segments, adds 20B header r IP adds 20B header, sends to network r If Ethernet, broken into 1500B frames with headers, trailers (24B) r All Headers, trailers have length field, destination,...


Download ppt "Computer Network Basics Components of Any Computer Processor (active) Computer Control (“brain”) Datapath (“brawn”) Memory (passive) (where programs,"

Similar presentations


Ads by Google