Lecture 3 #1

Chapter 3Hubs, Bridges and Switches

Lecture 3 #2

Interconnecting LANs

Q: Why not just one big LAN? Limited amount of supportable traffic: on

single LAN, all stations must share bandwidth Ethernet: limited length: 802.3 specifies

maximum cable length Ethernet: large “collision domain” (can collide

with many stations) Token Ring: token passing delay per station:

802.5 limits number of stations per LAN:

Lecture 3 #3

Hubs Physical Layer devices: essentially repeaters

operating at bit levels: repeat bits received on one interface to all other interfaces

Hubs can be arranged in a hierarchy (or multi-tier design), with backbone hub at its top

Lecture 3 #4

Hubs (more)

Each connected LAN referred to as LAN segment Hubs do not isolate collision domains: node may

collide with any node residing at any segment in LAN

Hub Advantages: simple, inexpensive device Multi-tier provides graceful degradation: portions

of the LAN continue to operate if one hub malfunctions

extends maximum distance between node pairs (100m per Hub)

Lecture 3 #5

Hub limitations

single collision domain results in no increase in max throughput multi-tier throughput same as single

segment throughput individual LAN restrictions pose limits on

number of nodes in same collision domain and on total allowed geographical coverage

cannot connect different Ethernet types (e.g., 10BaseT and 100baseT) Why?

Lecture 3 #6

Bridges Link Layer devices: forward Ethernet

frames selectively: learn where each station is located examine the header of each frame forward on the proper link (if known)

• if dest. and source on same link, drop frame WHY?

if not known where dest. is, broadcast frame• except on originating link, of course

Lecture 3 #7

Bridges Bridge isolates collision domains

buffers frame then forwards it, if needed, using CSMA/CD

A broadcast frame is forwarded on all interfaces (except the incoming one) thus broadcast frames propagate across

brodges A set of segments connected by bridges

and hubs is called a broadcast domain

Lecture 3 #8

Bridges (more)

Bridge advantages: Isolates collision domains resulting in higher

total max throughput, and does not limit the number of nodes nor geographical coverage

Can connect different type Ethernet since it is a store and forward device

Transparent: no need for any change to hosts LAN adapters

Lecture 3 #9

Backbone Bridge

Lecture 3 #10

Interconnection Without Backbone

Not recommended for two reasons:- single point of failure at Computer Science hub- all traffic between EE and SE must path over CS segment

Lecture 3 #11

Bridges: frame filtering, forwarding

bridges filter packets same-LAN -segment frames not forwarded

onto other LAN segments forwarding:

how to know on which LAN segment to forward frame?

Lecture 3 #12

Bridge Filtering

bridges learn which hosts can be reached through which interfaces: maintain filtering tables when frame received, bridge “learns” location

of sender: incoming LAN segment records sender location in filtering table

filtering table entry: (Node LAN Address, Bridge Interface, Time

Stamp) stale entries in Filtering Table dropped (TTL can

be 60 minutes)

Lecture 3 #13

Bridge Operation pseudocodeInit: set filtering table to voidCase: frame arrives on port P, src MAC , dest MAC

/* Table Update stage */ if not listed, add mapping P with expiration timeelse update expiration time /* if listing fits */

/* Frame Forwarding stage */ look up in filtering table: listing P Q /* if listed */if not listed, forward on all ports except P /* “flood */if = P , drop the frame /*frame already there*/

/* WHY ? */Otherwise, forward the frame on port Q

Lecture 3 #14

Bridge Learning: example

Suppose C sends frame to D and D replies back with frame to C

C sends frame, bridge has no info about D, so floods to both LANs 2 and 3 bridge notes that C is on port 1 frame ignored on upper LAN frame received by D

Lecture 3 #15

Bridge Learning: example

D generates reply to C, sends it bridge sees frame from D bridge notes that D is on interface 2 bridge knows C on interface 1, so selectively

forwards frame out via interface 1 only

C 1

Lecture 3 #16

What will happen with loops?Incorrect learning

A

B

1 1

22

A , 1 A , 122

Lecture 3 #17

What will happen with loops?Frame looping

A

C

1 1

22

C,?? C,??

……

Lecture 3 #18

Loop-free: tree

A

B

C

A message from Awill mark A’s location

Lecture 3 #19

Loop-free: tree

A

B

C

A message from Awill mark A’s location

A:

Lecture 3 #20

Loop-free: tree

A

B

CA:

A:

A message from Awill mark A’s location

Lecture 3 #21

Loop-free: tree

A

B

CA: A:

A:

A:

A:

A message from Awill mark A’s location

Lecture 3 #22

Loop-free: tree

A

B

CA: A:

A:

A:

A:

A message from Awill mark A’s location

Lecture 3 #23

Loop-free: tree

A

B

C

A:

A: A:

A:

A:

So a message toA will go by marks…

A message from Awill mark A’s location

Lecture 3 #24

Bridges-Spanning Tree for increased reliability, it is desirable to have

redundant, alternative paths from source to dest this causes cycles - bridges may multiply and

forward frame forever solution: organize bridges in a spanning tree and

disable all ports not aligned with the tree

Disabled

Lecture 3 #25

Introducing Spanning Tree Objective: Find tree spanning all LAN segments

each bridge transmits on a single port each LAN transmits on a single bridge

Bridges run the Spanning Tree Protocol Use a distributed algorithm Result: select what ports (and bridges) should

actively forward frames, and which should accept frames

bridges communicate using special configuration messages (BPDUs) to perform this selection• BPDU = Bridge Protocol Data Unit

STP standardized in IEEE 802.1D

Lecture 3 #26

Method Bridges communicate using special

configuration messages (BPDUs) BPDU = Bridge Protocol Data Unit

Each bridge sends periodically a BPDU to all its neighbors

BPDU contains: ID of bridge sender views as root (my_root_ID) known distance to that root senders own bridge ID

Lecture 3 #27

Overview of STP

We solve in order:1. How to agree on a root bridge?2. How to compute a ST of bridges?3. How to compute a ST LAN segments?

Lecture 3 #28

1. Choosing a root bridge Assume

each bridge has a unique identifier (ID) within a bridge each port has a unique ID

Each bridge remembers smallest bridge ID seen so far (= my_root_ID) including own ID

Periodically, send my_root_ID to all neighbors (“flooding”) (included in BPDU)

When receiving ID, update if necessary Qn: Is that enough?!

Lecture 3 #29

2. Compute ST given a root

Idea: each node finds its shortest path to the root shortest paths tree

Output: At each node, parent pointer (and distance)

How: Bellman-Ford algorithm

Lecture 3 #30

Distributed Bellman-Ford

Assumption: There is a unique root node s this was done in Step 1

Idea: Each node, periodically, tells all its neighbors what is its distance from s

But how can they tell? s: easy. dists = 0 always! Another node v: Bridge calls the neighbor with least

distance to root - its “parent”

Suppose all nodes start with distance , and suppose that updates are sent every time unit.

11

0

00

0

Lecture 3 #31

Why does this work?

CD

B

E

F

G

A 0

1 1

1

1

3

2

2

B sees same distance from A and E; A chosen since has smaller ID

ID=17

ID=21

2

2

Lecture 3 #32

Bellman-Ford: properties

Works for any non-negative link weights w(u,v):

Works when the system operates asynchronously.

Works regardless of the initial distances!

Lecture 3 #33

3. ST of LAN segmentsAssumption: given a ST of the bridgesIdea: Each segment has at least one bridge

attached. Only one of them should forward packets! Choose bridge closest to root. Break ties by bridge ID

(and then by port ID on that bridge if needed)

Implementation: Bridges listen to all distance announcements on each port. Bridge A marks a port as a “designated port” iff A is the best bridge for that port’s LAN, i.e: Its distance D to root is the shorter than all distances

received on this port Of all bridges with distance D reported on this port it

has lowest ID of several ports to same LAN on same bridge take low

ID

Lecture 3 #34

Spanning Tree Concepts:Path Cost A cost is associated with each segment

= “weight” of the segment = cost associated with transmission on the LAN

segment connected to the port bridge associates the weight with relevant port default segment weight is 1 Can be manually or automatically assigned Can be used to alter the path to the root bridge

Path cost is the sum of the component segment weights

Lecture 3 #35

Spanning Tree Concepts:Root Port Each non-root bridge has a Root port:

The port on the path towards the root bridge = parent pointer

The root port is part of the lowest cost path towards the root bridge

If port costs are equal on a bridge, the port with the lowest ID becomes root port

Lecture 3 #36

Example Spanning Tree

B3

B5

B7B2

B1

B6 B4

Protocol operation:1. Pick a root2. Each bridge picks a

root port

B8

Lecture 3 #37

Example Spanning Tree

B3

B5

B7B2

B1

B4B6

Root

B4 B5 B6

B8

B1

Spanning Tree:

root port

B3

B7B2

B8

Lecture 3 #38

ST Concepts: Designated Port Each LAN has a single designated bridge

all other bridges on LAN know which one it is all tfc of LAN towards root goes thru that bridge

This is the bridge reporting minimum cost path to the root bridge for the LAN ties broken by choosing lowest ID

Only designated & root ports remain active in a bridge. Bridge uses: designated ports to send downstream frames root ports to send upstream frames (toward root) Bridge with no designated port becomes inactive

Lecture 3 #39

Example Spanning Tree

B3

B5

B7B2

B1

B6 B4

Root

B8

B2 B4 B5 B7

B8

B1

Forwarding Tree:

Designated Bridge

root port

Note: B3, B6 forward nothing

Lecture 3 #40

STP Requirements

Each bridge has a unique identifierA multicast address for bridges on

a LANA unique port identifier for all ports

on all bridgesBridge id + port number

Lecture 3 #41

Forwarding/Blocking State

1. Only root and designated ports are active for data forwarding

Other ports are in the blocking state: no forwarding!

If bridge has no designated port, no forwarding at all block root port too.

2. All ports send BPDU messages including blocked ones To adjust to changes

Lecture 3 #42

Spanning Tree Protocol: Execution

B3

B5

B7B2

B1

B6 B4

B8

(B1,root=B1, dist=0)(B1,root=B1,dist=0)

(B4, root=B1, dist=1)(B6, Root=B1dist=1)

WHY?

Lecture 3 #43

Bridges vs. Routers both are store-and-forward devices

routers: network layer devices (examine network layer headers)

bridges are link layer devices

routers have routing tables, use routing algorithms, designed for Wide Area addressing

bridges have filtering tables, use filtering, learning & spanning tree algorithms, designed for local area

Lecture 3 #44

Routers vs. Bridges

Bridges + and - + Bridge operation is simpler, requiring less

processing- Topologies are restricted with bridges: a

spanning tree must be built to avoid cycles - Bridges do not offer protection from broadcast

storms (endless broadcasting by a host will be forwarded by a bridge)

Lecture 3 #45

Routers vs. Bridges

Routers + and -+ arbitrary topologies can be supported, cycling is

limited by TTL counters (and good routing protocols)+ provide firewall protection against broadcast storms- require IP address configuration (not plug and play)- require higher processing

bridges do well in small (few hundred hosts) while routers used in large networks (thousands of hosts) and in Internet core

Lecture 3 #46

Ethernet Switches

layer 2 (frame) forwarding, filtering using LAN addresses

Switching: A-to-B and A’-to-B’ simultaneously, no collisions

large number of interfaces often: individual hosts,

star-connected into switch Ethernet w. no

collisions! = Switched Ethernet

Lecture 3 #47

Ethernet Switches

cut-through switching: frame forwarded from input to output port without awaiting for assembly of entire frameslight reduction in latency

combinations of shared/dedicated, 10/100/1000 Mbps interfaces

Lecture 3 #48

Ethernet Switches (more)Dedicated

Shared

Lecture 3 #49

Data Link: Summary principles behind data link layer services:

error detection, optional: error correction sharing a broadcast channel: multiple access link layer addressing, ARP

various link layer technologies Ethernet hubs, bridges (STP), switches IEEE 802.11 LANs PPP

Chapter 5 Kurose and Ross

Lecture 3 #50

EXTRA SLIDES

Lecture 3 #51

Optional: Wireless LAN and PPP

Lecture 3 #52

IEEE 802.11 Wireless LAN wireless LANs: mobile networking IEEE 802.11 standard:

MAC protocol unlicensed frequency spectrum: 900Mhz,

2.4Ghz Basic Service Set (BSS) (a.k.a. “cell”) contains: wireless hosts access point (AP):

base station BSS’s combined to

form distribution system (DS)

Lecture 3 #53

Ad Hoc Networks Ad hoc network: IEEE 802.11 stations can

dynamically form network without AP Applications:

“laptop” meeting in conference room, car

interconnection of “personal” devicesbattlefield

IETF MANET (Mobile Ad hoc Networks) working group

Lecture 3 #54

IEEE 802.11 MAC Protocol: CSMA/CA802.11 CSMA: sender- if sense channel idle for

DIFS sec. then transmit entire frame

(no collision detection)-if sense channel busy

then binary backoff

802.11 CSMA receiver:if received OK return ACK after SIFS

Why Needed?

Lecture 3 #55

IEEE 802.11 MAC Protocol

802.11 CSMA Protocol: others

NAV: Network Allocation Vector

802.11 frame has transmission time field

others (hearing data) defer access for NAV time units

Lecture 3 #56

Hidden Terminal effect

hidden terminals: A, C cannot hear each other obstacles, signal attenuation collisions at B

goal: avoid collisions at B CSMA/CA: CSMA with Collision Avoidance

Lecture 3 #57

Collision Avoidance: RTS-CTS exchange CSMA/CA: explicit

channel reservation sender: send short

RTS: Request To Send receiver: reply with

short CTS: Clear To Send

CTS reserves channel for sender, notifying (possibly hidden) stations

avoid hidden station collisions

Lecture 3 #58

Collision Avoidance: RTS-CTS exchange

RTS and CTS short: collisions less likely, of

shorter duration end result similar to

collision detection IEEE 802.11 allows:

CSMA CSMA/CA: reservations polling from AP

Lecture 3 #59

Point to Point Data Link Control one sender, one receiver, one link:

easier than broadcast link:no Media Access Controlno need for explicit MAC addressinge.g., dialup link, ISDN line

popular point-to-point DLC protocols:PPP (point-to-point protocol)HDLC: High level data link control

(Data link used to be considered “high layer” in protocol stack!)

Lecture 3 #60

PPP Design Requirements [RFC 1557] packet framing: encapsulation of network-layer

datagram in data link frame carry network layer data of any network layer

protocol (not just IP) at same time ability to demultiplex upwards

bit transparency: must be able to carry any bit pattern in the data field

error detection (no correction) connection livenes: detect, signal link failure to

network layer network layer address negotiation: endpoint can

learn/configure each other’s network address

Lecture 3 #61

PPP non-requirements

no error correction/recovery no flow control out of order delivery OK no need to support multipoint links

(e.g., polling)

Error recovery, flow control, data re-ordering all relegated to Transport layer!!!

Lecture 3 #62

PPP Data Frame (1)

Flag: delimiter (framing) Address: does nothing (only one option) Control: does nothing; in the future possible

multiple control fields Protocol: upper layer protocol to which frame

delivered (eg, PPP-LCP, PPP-NCP, IP, IPCP, etc)

Lecture 3 #63

PPP Data Frame (2)

info: upper layer data being carried check: cyclic redundancy check (CRC)

for error detection

Lecture 3 #64

Byte Stuffing “data transparency” requirement: data field

must be allowed to include flag pattern <01111110> Q: is received <01111110> data or flag?

Byte Stuffing procedure Sender: adds (“stuffs”) extra < 01111101> byte

before each < 01111110> or <01111101> data byte

Receiver: when receive 01111101

• discard the byte, • Next byte is data, regardless of value

Receive 01111110: flag byte

Lecture 3 #65

Byte Stuffing

flag bytepatternin datato send

flag byte pattern plusstuffed byte in transmitted data

Lecture 3 #66

PPP Data Control ProtocolsBefore exchanging network-

layer data, data link peers must

PPP-LCP: configure PPP link (max. frame length, authentication)

learn/configure network layer information

for IP: carry IP Control Protocol (PPP-IPCP) msgs (protocol field: 8021) to configure/learn IP address