Docstoc

04 The OSI Network Layer

Document Sample
04 The OSI Network Layer Powered By Docstoc
					                        Revision no.: PPT/2K403/02




The OSI Network Layer
                                                                                                                                                                        Revision no.: PPT/2K403/02




OSI Network Layer


•   The Network layer protocol describes methods for moving
    information between multiple independent (and often
    dissimilar) networks, called internetworks. This function is
    commonly called routing.
•   The Network layer routes data using
•   Switching
•   Network layer addressing
•   Routing algorithms



     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




OSI Network Layer (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




OSI Network Layer Processes and Methods




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                      Revision no.: PPT/2K403/02




Addressing Methods




   © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Logical Network


•   To deliver data between networks on an internetwork, logical
    network addresses are used . A logical network address is the
    identifier used to logically (as opposed to physically)
    distinguish two networks in an internetwork.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Service Address


•   A service address identifies a specific upper-layer software

    process or protocol. Multiple service addresses can be

    assigned to any computer on which several network

    applications are running.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Service Address (contd.)


 •   Some networks reserve a bank of addresses, called well-
     known addresses, that identify common network services.
     Every network service provider can also supply its own unique
     service address.
 •   Service addresses to the logical network and physical device
     addresses:




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Service Address (contd.)


 •   The logical network address indicates the source or
     destination network.
 •   The physical device address identifies the source or
     destination computer.
 •   The service address refers to the specific application process
     running on the source or destination computer.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Service Address




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Switching Methods




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Circuit switching


 •   Circuit switching is a technique that connects the sender and
     the receiver by a single path for the duration of a conversation.
     After a connection is established, a dedicated path exists
     between both ends.
 •   Circuit switching in a computer network operates in the same
     way. The computer initiating the transfer asks for a connection
     to the destination.
 •   After the connection is made, the destination device
     acknowledges that it is ready to carry on a transfer.


      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Circuit switching (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Message Switching


•   Message switching does not establish a dedicated path
    between two workstations for an entire conversation. Rather,
    conversations are divided into message.
•   Message switching, or store-and-forward techniques, are used
    to support services such as email, calendaring, workflow, and
    groupware.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Message Switching (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Datagram Packet switching


•   Packets are switched as datagrams to the appropriate network

    and all devices on the same network segment receive each

    packet and determine whether to accept it using LLC-level

    addressing.

•   Datagram packet switching relies on the Network layer to

    navigate paths for each packet and to correct errors that might

    occur.


     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Datagram Packet switching (contd.)


 •   Datagram packet switching is common on the Internet. When
     you access a Web site that allows public access (doesn’t
     require a login), the data from the Web server might be sent to
     your computer using datagram packet switching.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Datagram Packet switching (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Route Discovery Methods




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Distance vector


 •   Distance vector routers compile and send network route tables

     to other routers that are attached to the same media segments.

 •   Distance vector router can only get update information from its

     neighbors. When this information is received, the router

     updates its routing table.

 •   The process of updating all routers’ routing tables so that they

     all contain the same data is called Convergence.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Link-state route


 •   Link-state route discovery protocols are considered more

     intelligent and faster than distance vector-based routing

     protocols.

 •   The link-state method only exchanges information about

     specific routes that have changed, rather than broadcasting

     route information periodically regardless of network

     conditions.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Link-state route (contd.)


 •   This information is used to create an area map that is used to

     determine the best route to send a packet.

 •   Router broadcasts this information, the receiving devices build

     a table of network addresses.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Route Selection Methods




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Routing


•   To forward packets from node to node, packet- and message-

    switched networks must constantly determine and use the

    correct path for each piece of data. This path determination

    task is called Routing.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Dynamic Route


•   Dynamic route selection uses routing algorithms to continually

    gather and evaluate cost information. Every packet is assigned

    a route depending on the latest route discovery costs.

•   Multiple paths might be used to send packets between two

    devices, depending on the changing nature of the network.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Dynamic Route (contd.)


 •   Dynamic route selection is flexible and can recover from failed

     or overworked routers. Routers that use dynamic route

     selection require little or no management, but the routing

     devices are complex and often expensive.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Static route


 •   Static route selection is based on a path designated by the

     network administrator or by an assigned network device (often

     the transmission’s source device).

 •   Static route selection can use simple and inexpensive routing

     devices, but this method often cannot recover from failed or

     overworked routers and can require a great deal of

     management.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Connection Services Methods




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Network layer flow control


 •   While Data Link layer flow control uses acknowledgments to

     control data flow based on device capabilities, Network layer

     flow control is based on the internetwork’s capabilities.

 •   Network layer flow control includes the mechanisms used to

     control the amount of data sent on a given route when multiple

     paths exist between sender and receiver.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Packet sequence control


•   Packet sequence control is used to put the arriving packets

    into proper sequence to rebuild upper-layer messages.

•   This type of control is required for datagram networks where

    packets routinely arrive out of order. Packet sequence control

    might also be necessary with large virtual circuit networks.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Error Control


 •   Error control is primarily concerned with packet loss, duplicate
     packets, and altered data.
 •   Packet loss errors are handled using acknowledgments in the
     same way as Data Link layer error control.
 •   Duplicate packet errors are typically handled by dropping all
     but one of the duplicated packets.
 •   Errors involving altered data are typically detected (and
     sometimes corrected) by appending a CRC or other checksum
     to the packet.


      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Gateway Services




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Gateway Services (contd.)


 •   Gateways can be implemented at any layer of the OSI model. A

     gateway is just a device or application that translates from one

     set of rules to another. However, most gateways are

     implemented at the higher OSI layers.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Gateway Services (contd.)


 •   A default gateway is a router all network packets are sent to if

     they are destined for servers or devices outside of the local

     network segment. The default gateway is also referred to as

     default router, or as simply gateway.

 •   The differences are typically resolved by using a gateway,

     which interprets and translates the rules on two separate

     networks.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Connectivity Devices


•   Every physical medium is limited by the number of devices

    that can be attached to a single cable, the distance a signal

    can travel on the medium before the signal is either too weak

    or out of phase to be intelligible, and the amount of bandwidth

    that is available for data transmission.

•   Repeaters and bridges help overcome these limitations.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Selecting Connectivity Devices




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridging Functionality


 •   Bridges are devices that increase the total throughput of a

     LAN by keeping frames on the local segment local and filtering

     frames that belong on other LAN segments.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Bridging Functionality (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Increased Total Throughput


•   Bridges increase usable bandwidth. Bridges filter traffic

    destined for remote LAN segments based on their hardware

    addresses, significantly increasing the usable throughput of a

    multi-segment LAN.

•   Bridges work independently of all upper-layer protocols and

    thus can forward frames from many different upper-layer

    protocols. Even protocols that cannot be dynamically routed

    can be bridged.

     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Increased Total Throughput (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Bridges


•   Bridges function up to the Data Link layer of the OSI model.

    Bridges increase the media distance and device limits on a

    LAN because the signal is filtered and repeated to the

    appropriate segment.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




The 80/20 Rule


•   When designing a bridged LAN, you should follow the 80/20
    rule. According to the rule, 80 percent of the traffic should be
    local and no more than 20 percent should be crossover traffic.
•   The top part of the following figure shows two physical
    segments connected by a repeater. Only one segment can be
    transmitted on at a time because everything is repeated on the
    other segment.
•   The bottom part of the following figure shows two physical
    segments connected by a transparent bridge.


     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Bridges and the 80/20 Rule




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Transparent Bridges


•   Transparent bridges, as defined by the IEEE 802.1d

    specification, are so named because a sending device is

    unaware that a target device might exist on a remote segment,

    or that one or more bridges might exist in the path. The bridge

    is transparent to the sending device.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Transparent Bridges (contd.)


 •   In the previous figure, device A transmits a frame addressed to

     device B. As far as A is concerned, B exists on the same

     physical segment.

 •   The bridge is the intelligent device that checks the destination

     MAC address and repeats, or forwards, the frame to segment

     S2.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Transparent Bridges (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




The Five Bridge Port States




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




The Five Bridge Port States (contd.)


 •   Transparent bridges are connected to media segments at two

     or more bridge ports. Each port can be thought of as a network

     adapter connected to its particular media segment.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




The Five Bridge Port States (contd.)


 •   Disabled: A port in this state does not participate in spanning

     tree operation forwarding or learning. This state is like a

     manual offline switch; it is entered into and left by bridge

     management action.

 •   Blocking: This is the initialization state of all bridges and the

     constant state of a backup bridge in a redundant bridge pair.

     Only bridge protocol frames addressed to the bridge multicast

     address are processed.

      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




The Five Bridge Port States (contd.)


 •   Listening: A port in this state is preparing for the learning and

     forwarding states. A timer is associated with being in this

     state; the timer is set to allow the network time to settle down

     during a topology change.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




The Five Bridge Port States (contd.)


 •   Learning: A port in the learning state has its frame forwarding

     capabilities temporarily disabled. A timer is also used in this

     state to allow additional settling down time before forwarding

     to avoid temporary bridging loops.

 •   Forwarding: This is the normal, active state for a designated

     bridge. Forwarding and learning capabilities are both enabled.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




The Port Relay Entity


 •   A frame addressed to LAN segment 2 is transmitted by LAN

     segment 1 to port 1. Because both ports are in the forwarding

     state, port 1 consults and the filtering database and the frame

     is forwarded to port 2 for transmission on LAN segment 2.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Bridge




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Building the Filtering Database Step 1


 •   Bridge B1 has just been turned on. Station A has data to

     transmit to station D. Station A builds a frame that identifies

     station D as the destination MAC address.

 •   Station A includes its own address in the frame’s source

     address field.

 •   When station A transmits the frame, bridge B1 copies the

     frame from segment S1. Bridge B1 reads the source address

     — workstation A in this case.

      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Building the Filtering Database Step 1 (contd.)


 •   Bridge B1 copied the frame from the port attached to segment

     S1, it enters workstation A in its filtering database as existing

     on segment S1.

 •   Bridge B1 has not yet determined where workstation D is

     located, it forwards the frame to all active ports except the one

     connected to segment S1(the segment from which the frame

     was copied).



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Building the Filtering Database Step 1 (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Building the Filtering Database Step 2




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Building the Filtering Database Step 2 (contd.)


 •   When workstation D responds to workstation A in the

     following figure, it builds a frame that identifies workstation A

     as the destination address. Workstation D transmits the frame.

     Bridge B1 copies the frame from segment




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Building the Filtering Database Step 2 (contd.)


 •   Bridge B1 reads workstation D as the source address and

     adds workstation D to the filtering database as existing on

     segment 2. Bridge B1 checks the destination address.

 •   Bridge B1 has previously detected that workstation A is on

     segment S1, it forwards the frame only to segment S1.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Building the Filtering Database Step 3




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Building the Filtering Database Step 3 (contd.)


 •   Suppose workstation C transmits a frame to workstation D as

     shown in Bridge B1 copies the frame from segment S2. Bridge

     B1 reads that workstation C is the source address. Because

     workstation C does not exist in the database, it is added.

     Bridge B1 reads the destination address.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Building the Filtering Database Step 3 (contd.)


 •   Workstation D exists in the database. Because workstation D

     is on the same segment as workstation C, bridge B1 filters

     (discards) the frame and does not forward it to other

     segments.

 •   Bridge B1 eventually learns where all devices are located. It

     then forwards frames only when necessary.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridging Loops Part 1


 •   Segments S1 and S2 are connected by redundant bridges B1

     and B2. Workstation A transmits a frame addressed to

     workstation B. Bridges B1 and B2 both copy the frame from

     segment S1 (step 1).

 •   Because no entry for workstation A exists in either of their

     databases, bridges B1 and B2 both add workstation A to their

     respective databases as existing on segment S1.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Bridging Loops Part 1 (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridging Loops Part 2


 •   Bridge is physically identical to the other, one bridge forwards

     the frame to segment S2 just slightly before the other.

 •   Bridge B1 forwards the frame to S2 just slightly before bridge

     B2 (step 2). Because bridge B2 has no way of knowing that

     bridge B1 transmitted the frame, bridge B2 copies the frame

     from S2.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridging Loops Part 2 (contd.)


 •   Bridge B2 checks its database and discovers that it had

     initially detected workstation A on segment S1. Because a

     bridge always uses the most recent information, bridge B2

     changes its database to reflect that A is now located on S2

     (step 3). Bridge B2 forwards the frame to S1.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridging Loops Part 2 (contd.)


 •   While B2 is adjusting its database, B1 has copied the original

     frame forwarded by B2 to S2 (step 4). Like bridge B2, bridge B1

     updates its database to reflect that workstation A is now

     located on S2 (step 5) and then forwards the frame to S1.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Bridging Loops Part 2 (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridging Loops, Part 3


 •   The cycle is repeated for the frames that were forwarded back

     to segment S1. Both bridges copy the frame, adjust their

     respective databases to reflect the change, and then forward

     the frame back to segment S2.

 •   This happens for every frame appearing on both segments.

     The bridges and the LAN become saturated with traffic,

     making the LAN inoperable.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Bridging Loops Part 3 (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Source-Routing Bridges


•   Source-routing bridges, unlike transparent bridges, do not

    maintain a filtering database. Rather, each device in a source-

    routing network maintains its own dynamic table of routes to

    devices it communicates with. Hence the name source routing

    bridge.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Source-Routing Bridges (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Switching Hubs


•   Bridging has a number of limitations, including loops, a lack of

    redundancy, and message flooding.

•   Moving to a higher-performance LAN usually requires the

    replacement of network boards, hubs, routers and bridges,

    network printer interfaces, and, sometimes, even the wiring

    plant. This is all very costly in terms of hardware and time.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Switching Hubs (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Ethernet Switch Functionality


 •   The switch looks at the MAC address of every incoming frame

     and passes each frame only to the port on which its

     destination devices reside.

 •   A typical switch can monitor traffic on every port and

     concurrently switch among multiple ports.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Ethernet Switch Functionality (contd.)


 •   Each port can attach to a single high-volume user or server or

     a multiport 10BASE-T hub.

 •   The devices connected to each port define a collision domain

     where devices compete for bandwidth only with other devices

     on the same port, significantly reducing congestion and

     collisions on the network as a whole.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Ethernet Switch Functionality




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Virtual LANs


•   A switched virtual LAN is a broadcast domain that unites any

    arbitrary group of LAN segments at wire speed. As is the case

    with a single physical wire, broadcasts travel to all end-

    stations in a virtual LAN.

•   Some departmental switches handle virtual LANs at the Data

    Link layer, leaving Network layer functions to routers. Other

    switches handle virtual LANs at the Network layer, performing

    basic routing chores.

     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Virtual LANs (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Token Ring Switching


•   Token ring switch manufacturers use the same switching

    technologies used for Ethernet: cut-through (on-the-fly) and

    store-and-forward.

•   Switching and full-duplex token ring have not been needed

    until recently because most users have been satisfied with the

    16 Mbps bandwidth and high utilization inherent to regular

    token ring technology.



     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Token Ring Switching (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Inter-network Routing


 •   Routing protocols and routing enable devices to communicate

     with one another across an inter-network.

 •   To route packets across an inter-network, routers need a

     destination address and a known path to the destination.

     Routing is independent of the underlying media and uses

     software addresses.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Inter-network Routing (contd.)


 •   Software addresses differ from physical addresses. A physical

     address identifies a single device. A software address

     identifies the device and the network. The network portion of

     the address can reflect a hierarchy of networks.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Inter-network Routing (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Router R1’s Functionality


 •   Router R1 copies the frame from network 11 (step 1). The Data

     Link layer on R1 strips the header and trailer after verifying the

     destination address and the CRC. Everything else is passed up

     to the Network layer (step 2).

 •   R1’s ACME protocol examines the destination network portion

     of the software address and determines that the packet is

     destined for another network (network 33).



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Router R1’s Functionality (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Router R1’s Functionality (contd.)


 •   R1’s ACME protocol checks its routing information table (step

     3) to see which routers on a directly connected link can

     forward the packet to network 33. R1’s routing information

     table (a table of routes and addresses for devices on the

     network) indicates that network 33 is only one router, or hop,

     away.


      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Router R1’s Functionality (contd.)


 •   This means that router R2 is directly connected to the

     destination network. R1’s ACME protocol passes the packet

     back down to the Data Link layer with instructions to transmit

     the packet to router R2 (step 4).




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Router R1’s Functionality


 •   R1’s Data Link layer creates the token ring frame and transmits

     the packet onto network 22 (step 5).

 •   Router R2 copies the frame from network 22 (see the following

     figure). R2’s ACME protocol receives the packet from the Data

     Link layer, checks the network portion of the software address,

     and determines that it is directly connected to the destination

     network.



      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Router R2 Forwarding the Packet to Station CC


 •   Router R2 then checks the device portion of the software

     address to determine which end station should receive the

     packet. The ACME protocol passes the packet back down to

     the Data Link layer with instructions to address the packet to

     workstation CC.

 •   The frame now contains the hardware address of the

     destination station. Router R2 then transmits the frame on

     network 33.

      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Router R2 Forwarding the Packet to Station CC (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Building the Routing Database


•   Routers maintain databases known as routing information

    tables. As shown in the following figure, these tables tell the

    router the location of every network on the inter-network with

    respect to the router’s position in the inter-network.

•   When a router receives a packet, it forwards the packet to the

    target device or to another router.

•   The next router then forwards the packet to another router

    until the packet can be delivered to the destination device.

     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




The Routing Database




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Distance Vector Routing


•   A distance vector router builds its routing information table by
    calculating routes to all networks based on information it
    receives from neighboring routers.

•   If a router indicates it can reach network x in 3 hops, then that
    router’s neighbor assumes network x is 4 hops away. This
    figure is entered into the routing information table.

•   A distance vector router, in essence, uses second-hand
    information. The Routing Information table is based on
    information received from other routers.
     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Distance Vector Routing (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Count-to-Infinity Problem


 •   Infinity is actually the limit of the cost counter, which is
     determined by the implementation and typically configured by
     management software.

 •   The maximum hop count is commonly set to 16.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Count-to-Infinity Problem (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Link State Routing


 •   Link state routers build and transmit special packets called
     Link State Packets (LSPs), which contain information about a
     router’s directly connected links and their costs.

 •   The packet is flooded to other routers, which means that it is
     forwarded to all other networks except the one on which it was
     received.

 •   After a router has a copy of the LSP for all other routers, it has
     first-hand information for a reliable map of the inter-network.


      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Link State Routing (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Load balancing


•   Load balancing allows efficient use of available bandwidth.

•   Load balancing results in packets moving over slower routes,

    but increases overall bandwidth usage.

•   Load balancing is an option for both distance vector and link

    state routers.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Load balancing (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Non-routable Protocols


•   Usually non-routable means that a protocol uses only static
    routes that are pre-defined and cannot be updated. Non-
    routable protocols perform route discovery through a simple
    table lookup.

•   In other cases, non-routable protocols do not use a Network
    layer for routing.




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Non-routable Protocols (contd.)


 •   The following are non-routable protocols:

 •   Routing SNA:

 •   NetBIOS                          :                      Network Basic Input Output System

 •   NetBEUI                          :                      Network Basic Extended User Interface

 •   LAT                              :                      Local Area Transport.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Non-routable Protocols (contd.)




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridges, Switches, and Routers


 •   Bridges and switches are designed to bridge LAN segments
     and provide more usable bandwidth within a network.

 •   Routers were created to connect networks into inter-networks.
     Routers are used to limit broadcast traffic and to provide
     security, control, and redundancy between individual
     broadcast domains.

 •   A router has information about the entire inter-network and
     can determine the best path between itself and the destination
     device.
      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridges, Switches, and Routers (contd.)


 •   Bridges and switches work in a limited topology, as defined by
     a single network number. (However, some switches allow
     virtual LANs.)

 •   Routed inter-networks, especially those using link state
     routing, converge more quickly than spanning-tree networks.
     Routers continue to route packets as they alert other routers
     on the network of a change.




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                         Revision no.: PPT/2K403/02




Bridges, Switches, and Routers (contd.)


 •   A bridged network reconfigures itself more slowly, and all
     bridges must stop forwarding frames for an amount of time
     specified by the root bridge. Then the bridge have to relearn
     the locations of all devices before they can resume forwarding
     frames .




      © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                       Revision no.: PPT/2K403/02




Bridges, Switches, and Routers (contd.)




    © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                        Revision no.: PPT/2K403/02




Exercise

•   Exercise 4.4 Comparing Bridges, Switches, and Routers
•   Exercise 4.4Selecting Connectivity Devices for a Network




     © CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
                                                                                                                                                                   Revision no.: PPT/2K403/02




                                                    Design & Published by:
      CMS Institute, Design & Development Centre, CMS House, Plot No. 91, Street No.7,
              MIDC, Marol, Andheri (E), Mumbai –400093, Tel: 91-22-28216511, 28329198
                                            Email: courseware.inst@cmail.cms.co.in
                                                           www.cmsinstitute.co.in
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute

				
DOCUMENT INFO
Shared By:
Categories:
Tags: network, layer
Stats:
views:11
posted:1/20/2010
language:English
pages:110