Ch. 15 LAN Overview

1
Ch. 15 LAN Overview
2
Definition of a LAN

• A communication network that provides
  interconnection of a variety of data
  communicating devices within a small area.

3
15.1 Topologies and Transmission Media

• Key Elements of a LAN
• Topology
• Transmission Media
• Layout
• Medium access control

4
15.2 Topologies and Transmission Media(p.2)

• Bus and Tree Topologies (Fig. 15.1)
• Bus
• All stations are attached directly to the media.
• Tree
• The media is a branching cable with no closed
  loops.
• The tree starts at the headend and branches out
  from there.
• Each station must have an address and access is
  controlled (multipoint configuration.)Fig.15.2

5
15.2 Topologies and Transmission Media(p.3)

• Ring Topology (Fig. 15.3)
• Network consists of a set of repeaters joined by
  point-to-point links in a closed loop.
• The links are unidirectional, and data circulates
  around the ring in one direction.
• Each station is attached to a repeater, and
  frames are inserted onto the ring.

6
15.2 Topologies and Transmission Media (p.4)

• Star Topology
• Each station is connected to a common central
  node using two point-to-point links.
• Received frames can either be "broadcast" or
  "switched" to a particular link.

7
15.2 Topologies and Transmission Media (p.5)

• Choice of Topology
• Depends on reliability, expandability, and
  performance.
• Choice of Media
• Depends on capacity, reliability, type of data
  supported, environmental scope.

8
15.2 LAN Protocol Architecture

• Fig. 15.4 IEEE 802 vs. OSI Reference Model.
• Physical Layer
• Encoding/decoding of signals.
• Preamble generation/removal (for
  synchronization).
• Bit transmission/reception.
• IEEE 802 also specifies the transmission medium
  and topology.

9
15.2 LAN Protocol Architecture (p.2)

• Medium Access Control (MAC) Layer
• Assemble data into a frame with address and
  error-detection fields.
• Disassemble frames, perform address recognition
  and error detection
• Govern access to the LAN transmission medium.

10
15.2 LAN Protocol Architecture (p.3)

• Logical Link Control (LLC) Layer
• Provide an interface to higher layers and perform
  flow and error control.
• Fig. 15.5 LAN protocols in context.

11
15.2 LAN Protocol Architecture (p.4)

• Logical Link Control
• Specifies the mechanisms for addressing and the
  control of the data exchange.
• Operation and format are based on HDLC.
• Three Services
• Unacknowledged connectionless service.
• Connection-mode service.
• Acknowledged connectionless service.

12
15.2 LAN Protocol Architecture (p.5)

• Logical Link Control (cont.)
• LLC PDU (Fig. 15.6)
• Destination Service Access Point (1 octet)
• 7 bits for the address.
• One bit to indicate if it is a group address or
  not.
• Source Service Access Point (1 octet)
• 7 bits for the address.
• One bit is used to indicate if it is a command or
  response.
• LLC Control Field (1 or 2 octets)
• Similar to HDLC control field.
• Information Field (variable length)

13
15.2 LAN Protocol Architecture (p.6)

• Differences between LLC and HDLC
• LLC uses asynchronous balanced mode to support
  connection-mode service (type 2 operation).
• LLC supports and unacknowledged connectionless
  service using the unnumbered information PDU
  (type 1 service).
• LLC supports an acknowledged connectionless
  service by using two new unnumbered PDUs (type 3
  operation.)
• LLC permits multiplexing (using LSAPs).

14
15.2 LAN Protocol Architecture (p.7)

• Medium Access Control
• MAC protocols control access to the transmission
  medium in some type of orderly and efficient
  manner.
• Access control could be centralized or
  distributed.
• Centralized schemes tend to be simpler and avoid
  various "distributed control" problems, but
  performance and reliability can be a concern.

15
15.2 LAN Protocol Architecture (p.8)

• Medium Access Control (cont.)
• Synchronous Techniques
• Specific capacity is dedicated to a connection,
  such as with circuit-switching, FDM, and TDM.
• Generally do not work well in LANs.

16
15.2 LAN Protocol Architecture (p.9)

• Medium Access Control (cont.)
• Asynchronous techniques--capacity is allocated in
  a dynamic fashion.
• Round Robin--each station is given a turn to
  transmit.
• Reservation--a station wishing to transmit
  "reserves" slots of "time".
• Contention--all stations "contend" for the medium.

17
15.2 LAN Protocol Architecture (p.10)

• Medium Access Control (cont.)
• Generic MAC Frame Format--Fig. 15.6
• MAC Control Field
• Destination MAC Address
• Source MAC Address
• LLC PDU
• CRC

18
Problem 15.3

• Consider the transfer of a file containing one
  million 8-bit characters from one station to
  another. What is the total elapsed time and
  effective throughput for the following cases?
• a. Circuit-Switched LAN
• TtotalSwitchS L/Btprop
• ThroughputSwitch L/TtotalSwitch

19
Problem 15.3 (p.2)

• b. Bus Topology
• D--distance between stations.
• B--data rate (use R bps if you wish.)
• P--packet size.
• Header is 80 bits.
• Information field is P-80.
• Acknowledgement is 88bits.
• v200 m/microsecond.

20
Problem 15.3 (p.3)

• b. Bus Topology (cont.)
• Assume that each packet is acknowledge before the
  next is sent (stop-and-wait.)
• Let NoPa the number of packets.
• NoPa L/(P-80), rounded up (assuming fixed length
  packets and L is the number of inoformation bits
  in the message.)
• There will be NoPa cycles needed to transfer the
  entire message.

21
Problem 15.3 (p.4)

• b. Bus Topology (cont.)
• Ignore additional overhead--then tframeP/B.
• Also let tprop D/v and tack88/B.
• Then TcycleBustframe tproptacktprop (ignoring
  processing delays.)
• Thus, TtotalBusNoPa (TcycleBus)
• ThroughputBusL/TtotalBus

22
Problem 15.3(p.5)

• c. Ring Topology
• Total circular length is 2D, with the two
  stations a distance D apart.
• Acknowledgement occurs with the circulation of
  the packet past the destination station, back to
  the source station.
• There are N repeaters, each introduces a delay of
  one bit time (1/B).

23
Problem 15.3 (p.6)

• c. Ring Topology (cont.)
• Assume similar overhead as in part b.
• RingPropTime2D/v N/B
• TcycleRingtframeRingPropTime
• TtotalRingNoPa(TcycleRing)
• ThroughputRingL/TtotalRing

24
15.3 Bridges

• Bridges were originally used to interconnect LANs
  using the same physical and MAC protocols.
• Eventually, bridges were developed that
  interconnected LANs with different MAC protocols.
• In general, bridges are simpler than routers.

25
Bridge Operation

• Why use a bridge, instead of simply operating as
  one large LAN?
• Reliability--bridges can be used to partition a
  large LAN environment.
• Performance--in general, as stations are added to
  a LAN, the performance decreases.
• Security--different types of traffic with
  different security needs can be kept on
  physically separate media.
• Geography--two LANs in different locations can be
  bridged using point-to-point communications.

26
Functions of a Bridge

• See Fig. 15.7
• The bridge reads all frames transmitted on
  network A, accepting those addressed to B.
• Frames accepted are transmitted on B.
• The same is done for B-to-A traffic.

27
Design Considerations

• 1. The bridge makes no modifications to the
  content or format of the frames it receives.
• 2. The bridge should contain enough buffer space
  to meet peak demands.
• 3. The bridge must contain addressing and routing
  intelligence.
• 4. A bridge may connect more than two LANs.
• Note Bridges can be more complex and have
  special functionality

28
Bridge Protocol Architecture

• The IEEE 802 committee has produced
  specifications for bridges.
• These devices are called MAC-level relays.
• Fig. 15.8 illustrates the architecture and
  operation.

29
Routing with Bridges

• Figure 15.9 illustrates the concept of alternate
  routes.
• Three Strategies
• Fixed Routing
• Spanning Tree (IEEE 802.1)
• Source Routing (IEEE 802.5)

30
Routing with Bridges (p.2)

• Fixed Routing
• A route is selected for each source-destination
  pair of LANs in the internet.
• If alternative routes exist, then the route with
  the fewest hops in chosen and placed in a routing
  table.
• Widely used simple and requires minimal
  processing.
• Too limited for a dynamically changing internet.

31
Routing with Bridges (p.3)

• The Spanning Tree Approach
• Three mechanisms
• Frame Forwarding
• Address Learning
• Loop Resolution

32
Routing with Bridges (p.4)

• The Spanning Tree Approach (cont.)
• Frame Forwarding
• The bridge maintains a forwarding database for
  each port attached to a LAN.
• The database indicates the station addresses for
  which frames should be forwarded through that
  port.

33
Routing with Bridges (p.5)

• The Spanning Tree Approach (cont.)
• Address Learning
• When a frame arrives at a particular port, the
  source address can be checked.
• If the source address is not in the database for
  that port it can be added.
• Each time an element is added to the database, a
  timer can be set.
• When the timer expires, then the element will be
  removed from the database.
• If the element is already in the database, the
  timer is reset.

34
Routing with Bridges (p.6)

• The Spanning Tree Approach (cont.)
• Spanning Tree Algorithm--Loop Problems
• The above procedures work fine when the topology
  is a tree, but problems occur when alternate
  routes exist.
• Consider Fig. 15.10.
• When A transmits to B, both bridges will update
  their databases and relay the frame.
• However, they will receive each others relay and
  update the databases again.
• B then cannot transmit to A.

35
15.3 Routing with Bridges (p.7)

• The Spanning Tree Approach (cont.)
• Spanning Tree Algorithm--Some Assumptions
• 1.Each bridge is assigned a unique identifier.
• 2.There is a special group MAC address that means
  "all bridges on this LAN".
• 3. Each port of a bridge is uniquely identified
  within the bridge.
• These assumptions allow the bridges to exchange
  routing information in order to obtain a spanning
  tree.

36
15.4 Hubs and Switches

• Hubs
• The active central element of a star layout.
• Each station is connected to the hub with two
  lines, one for transmitting and one for
  receiving.
• The system is essential a logical bus, since a
  transmission from any one station is transmitted
  to all other stations.
• Multiple levels of hubs are possible (Fig.
  15.11.)
• Hubs are usually placed in a wiring closet.
• Stations are about 100 meters away, using twisted
  pair, or 500 meters with optical fiber.

37
15.4 Hubs and Switches (p.2)

• Layer 2 Switches (Fig. 15.12)
• A shared medium hub (like a shared medium bus)
  has collisions when more than one station is
  transmitting at the same time.
• A layer 2 switch takes an incoming frame and
  transmits it only on the destination stations
  line.
• Two types of switches
• Store-and-Forward--packets are buffered.
• Cut-through--headers are read and switching
  occurs immediately--but no error checking.

38
15.4 Hubs and Switches (p.3)

• Layer 2 switches may function as a multiport
  bridge--the differences are
• Bridge frames are handled in software, while
  layer 2 switches have hardware that performs
  address recognition and frame forwarding.
• A bridge handles one frame at a time, while a
  switch can handle multiple frames at a time.
• A bridge uses store and forward operations, while
  cut-through operations are possible with layer 2
  switches.

39
15.5 Virtual LANS

• Figure 15.13, page 469 illustrates a typical LAN
  configuration.
• Consider a single MAC frame from X.
• Assume that X wants to transmit to Ythe local
  switch transmits it to Y.
• Alternatively, assume that X wants to transmit to
  W or Zthen the local switch routes the frame
  accordinglyunicast addressing.

40
VLANS (p.2)

• Broadcasting is also possible using a broadcast
  address.
• One approach to efficient transmissionpartition
  the LAN into separate broadcast domains.
• Figure 15.14 illustrates the use of a router for
  partitioning a LANIP addresses are used for
  routingthis may not be efficient either.

41
The Use of VLANs

• VLAN logic is implemented in LAN switches and
  functions at the MAC layer.
• A VLAN is a logical subgroup within a LAN that is
  created by software rather than by physical
  partitioning.
• Figure 15.15 illustrates a VLAN Configuration.

42
VLANS (cont.)

• From a business view, the VLAN provides the
  ability to be physically dispersed while
  maintaining its group identity.

43
Defining VLANs

• A VLAN is a broadcast domain consisting of a
  group of end stations that are not constrained by
  their physical locations.
• Approaches
• Membership by Port Group
• Membership by MAC Address
• Membership based on Protocol Information

44
Membership by Port Group

• Each switch has two types of ports.
• Trunk ports will connect switches and end ports
  will connect workstations to the switch.
• A VLAN can be defined by assigning each end port
  to a particular VLAN
• Advantageeasy to configure.
• DisadvantageNetwork manager must take care of
  configurations manually.

45
Membership by MAC Address

• MAC Addresses on in the hardware network
  interface cards (NICs).
• If a network manager physically moves a machine,
  the device automatically retains its VLAN
  membership.
• DisadvantageVLAN membership is assigned
  initially, which is difficult in large
  organizations. There is also a problem when
  docking stations are usedthey contain the NICs.

46
Membership Based on Protocol Information

• IP addresses can be used to assign VLAN
  membership.
• Or, transport protocol information could be used
  (or even higher protocol information.)
• Advantageflexible.
• Disadvantageissues related to performane and the
  processing of MAC addresses and other addressing.