Chapter 7: Network Architectures

1
Chapter 7Network Architectures
2
Learning Objectives

• Understand the different major network
  architectures, including 10 Mbps Ethernet, 100
  Mbps Ethernet, Gigabit Ethernet, token ring,
  AppleTalk, FDDI, and ATM
• Understand the standards governing network
  architectures
• Understand the limitations, advantages, and
  disadvantages of each standard or architecture

3
Ethernet

• Many experiments in early 1960s and 1970s to
  connect several computers and share data
• ALOHA network at University of Hawaii
• Early version of Ethernet developed at Xeroxs
  Palo Alto Research Center in 1972
• DIX (Digital, Intel, Xerox) developed standard
  that transferred at 10 Mbps
• IEEE used it as basis for 802.3 specification

4
Overview of Ethernet

• Popular network architecture with many
  advantages
• Ease of installation
• Low cost
• Support for different media
• Features include packaging data into frames,
  using CSMA/CD channel access, and using hardware
  (MAC) address
• Divided into three categories based on
  transmission, speed, and media

5
Ethernet (CSMA/CD)

• Carriers Sense Multiple Access with Collision
  Detection
• Xerox - Ethernet
• IEEE 802.3

6
IEEE802.3 Medium Access Control

• Random Access
• Stations access medium randomly
• Contention
• Stations content for time on medium

7
ALOHA

• Packet Radio
• When station has frame, it sends
• Station listens (for max round trip time)plus
  small increment
• If ACK, fine. If not, retransmit
• If no ACK after repeated transmissions, give up
• Frame check sequence (as in HDLC)
• If frame OK and address matches receiver, send
  ACK
• Frame may be damaged by noise or by another
  station transmitting at the same time (collision)
• Any overlap of frames causes collision
• Max utilization 18

8
Slotted ALOHA

• Time in uniform slots equal to frame transmission
  time
• Need central clock (or other sync mechanism)
• Transmission begins at slot boundary
• Frames either miss or overlap totally
• Max utilization 37

9
CSMA

• Propagation time is much less than transmission
  time
• All stations know that a transmission has started
  almost immediately
• First listen for clear medium (carrier sense)
• If medium idle, transmit
• If two stations start at the same instant,
  collision
• Wait reasonable time (round trip plus ACK
  contention)
• No ACK then retransmit
• Max utilization depends on propagation time
  (medium length) and frame length
• Longer frame and shorter propagation gives better
  utilization

10
Nonpersistent CSMA

• If medium is idle, transmit otherwise, go to 2
• If medium is busy, wait amount of time drawn from
  probability distribution (retransmission delay)
  and repeat 1
•  Random delays reduces probability of collisions
• Consider two stations become ready to transmit at
  same time
• While another transmission is in progress
• If both stations delay same time before retrying,
  both will attempt to transmit at same time
• Capacity is wasted because medium will remain
  idle following end of transmission
• Even if one or more stations waiting
• Nonpersistent stations deferential

11
1-persistent CSMA

• To avoid idle channel time, 1-persistent protocol
  used
• Station wishing to transmit listens and obeys
  following 
• If medium idle, transmit otherwise, go to step 2
• If medium busy, listen until idle then transmit
  immediately
• 1-persistent stations selfish
• If two or more stations waiting, collision
  guaranteed
• Gets sorted out after collision

12
P-persistent CSMA

• Compromise that attempts to reduce collisions
• Like nonpersistent
• And reduce idle time
• Like1-persistent
• Rules
• If medium idle, transmit with probability p, and
  delay one time unit with probability (1 p)
• Time unit typically maximum propagation delay
• If medium busy, listen until idle and repeat step
  1
• If transmission is delayed one time unit, repeat
  step 1
• What is an effective value of p?

13
Value of p?

• Avoid instability under heavy load
• n stations waiting to send
• End of transmission, expected number of stations
  attempting to transmit is number of stations
  ready times probability of transmitting
• np
• If np gt 1 on average there will be a collision
• Repeated attempts to transmit almost guaranteeing
  more collisions
• Retries compete with new transmissions
• Eventually, all stations trying to send
• Continuous collisions zero throughput
• So np lt 1 for expected peaks of n
• If heavy load expected, p small
• However, as p made smaller, stations wait longer
• At low loads, this gives very long delays

14
CSMA/CD

• With CSMA, collision occupies medium for duration
  of transmission
• Stations listen whilst transmitting
• If medium idle, transmit, otherwise, step 2
• If busy, listen for idle, then transmit
• If collision detected, jam then cease
  transmission
• After jam, wait random time then start from step 1

15
CSMA/CDOperation
16
Which Persistence Algorithm?

• IEEE 802.3 uses 1-persistent
• Both nonpersistent and p-persistent have
  performance problems
• 1-persistent (p 1) seems more unstable than
  p-persistent
• Greed of the stations
• But wasted time due to collisions is short (if
  frames long relative to propagation delay
• With random backoff, unlikely to collide on next
  tries
• To ensure backoff maintains stability, IEEE 802.3
  and Ethernet use binary exponential backoff

17
Binary Exponential Backoff

• Attempt to transmit repeatedly if repeated
  collisions
• First 10 attempts, mean value of random delay
  doubled
• Value then remains same for 6 further attempts
• After 16 unsuccessful attempts, station gives up
  and reports error
• As congestion increases, stations back off by
  larger amounts to reduce the probability of
  collision.
• 1-persistent algorithm with binary exponential
  backoff efficient over wide range of loads
• Low loads, 1-persistence guarantees station can
  seize channel once idle
• High loads, at least as stable as other
  techniques
• Backoff algorithm gives last-in, first-out effect
• Stations with few collisions transmit first

18
Collision Detection

• On baseband bus, collision produces much higher
  signal voltage than signal
• Collision detected if cable signal greater than
  single station signal
• Signal attenuated over distance
• Limit distance to 500m (10Base5) or 200m
  (10Base2)
• For twisted pair (star-topology) activity on more
  than one port is collision
• Special collision presence signal

19
IEEE 802.3 Frame Format
20
10 Mbps IEEE Standards

• Four major implementations
• 10Base5 using thick coaxial cable
• 10Base2 using thinnet coaxial cable
• 10BaseT using unshielded twisted-pair (UTP)
  cable
• 10BaseF using fiber-optic cable
• Of these standards only 10BaseT and 10BaseF are
  commonly seen today

21
10BaseT

• Uses Category 3, 4, or 5 unshielded twisted-pair
  (UTP) cable
• Low cost makes it most popular Ethernet network
• Wired as star topology but uses bus signaling
  system internally, as shown in Figure 7-1
• No more than five cabling segments, no more than
  four hubs between communicating workstations
• Up to 1024 computers

22
10BaseT Network Uses Star Topology
23
10BaseT (continued)

• 100 meter maximum cable segment length
• Table 7-1 summarizes 10BaseT Ethernet
• See Simulation 7-1 for a visual study of Ethernet
  operation

24
10BaseT Ethernet Summary
25
10BaseF

• Uses fiber-optic cable
• Three subcategories
• 10BaseFL links computers in LAN environment
• 10BaseFP links computers using passive hubs
  maximum cable segment length of 500 meters
• 10BaseFB uses fiber-optic cable as backbone
  between hubs
• Usually wired as a star with maximum of 1024
  nodes connected by repeaters
• Table 7-2 summarizes 10BaseF Ethernet

26
10BaseF Ethernet Summary
27
100 Mbps IEEE Standards

• Two most popular 100 Mbps Ethernet standards are
• 100BaseT, also called Fast Ethernet
• 100 VG-AnyLAN Short-lived technology that is
  rarely if ever seen in todays networks

28
100BaseT

• Current IEEE standard is 802.3u
• Three substandards define cable type
• 100BaseT4 four-pair Category 3, 4, or 5 UTP
• 100BaseTX two-pair Category 5 UTP
• 100BaseFX two-strand fiber-optic cable

29
100BaseT (continued)

• Two types of 100BaseT hubs
• Class I may have only one between communicating
  devices
• Class II may have maximum of two between
  devices
• Figure 7-2 shows switches interconnecting
  multiple hubs
• Table 7-3 summarizes 100BaseT Ethernet

30
Switch Interconnects 100BaseT Hubs
31
Summary of 100BaseT Ethernet
32
Gigabit Ethernet Physical

• 1000Base-SX
• Short wavelength, multimode fiber
• 1000Base-LX
• Long wavelength, Multi or single mode fiber
• 1000Base-CX
• Copper jumpers lt25m, shielded twisted pair
• 1000Base-T
• 4 pairs, cat 5 UTP
• Signaling - 8B/10B

33
Gbit Ethernet Medium Options(log scale)
34
10Gbps Ethernet - Uses

• High-speed, local backbone interconnection
  between large-capacity switches
• Server farm
• Campus wide connectivity
• Enables Internet service providers (ISPs) and
  network service providers (NSPs) to create very
  high-speed links at very low cost
• Allows construction of (MANs) and WANs
• Connect geographically dispersed LANs between
  campuses or points of presence (PoPs)
• Ethernet competes with ATM and other WAN
  technologies
• 10-Gbps Ethernet provides substantial value over
  ATM

35
10Gbps Ethernet - Advantages

• No expensive, bandwidth-consuming conversion
  between Ethernet packets and ATM cells
• Network is Ethernet, end to end
• IP and Ethernet together offers QoS and traffic
  policing approach ATM
• Advanced traffic engineering technologies
  available to users and providers
• Variety of standard optical interfaces
  (wavelengths and link distances) specified for 10
  Gb Ethernet
• Optimizing operation and cost for LAN, MAN, or
  WAN 

36
10Gbps Ethernet - Advantages

• Maximum link distances cover 300 m to 40 km
• Full-duplex mode only
• 10GBASE-S (short)
• 850 nm on multimode fiber
• Up to 300 m
• 10GBASE-L (long)
• 1310 nm on single-mode fiber
• Up to 10 km
• 10GBASE-E (extended)
• 1550 nm on single-mode fiber
• Up to 40 km
• 10GBASE-LX4
• 1310 nm on single-mode or multimode fiber
• Up to 10 km
• Wavelength-division multiplexing (WDM) bit stream
  across four light waves

37
10Gbps Ethernet Distance Options (log scale)
38
Gigabit Ethernet 1 Gbps IEEE 802.3z Standards

• 1000BaseX identifies various Gigabit Ethernet
  standards
• Requires different signaling methods
• Uses 8B/10B coding scheme with 8 bits of data and
  2 bits of error-correction data
• Most use full-duplex mode

39
Gigabit Ethernet 1 Gbps IEEE 802.3z Standards
(continued)

• Two separate extensions cover 1000BaseX and
  1000BaseT
• 802.3z-1998 covers 1000BaseX including
• L long wavelength laser/fiber-optic
• S short wavelength laser/fiber-optic
• C copper jumper cables
• 802.3ab-1999 covers 1000BaseT requiring four
  pairs of 100-ohm Category 5 cable or better

40
10 Gigabit Ethernet10 Gbps IEEE 802.3ae Standard

• Anticipated ratification in late 2002
• Runs only on fiber-optic cabling, using both
  single-mode and multi-mode
• Maximum length is 5 km
• Uses full-duplex
• Likely to be used as network backbone and in
  Storage Area Networks (SANs)
• Able to scale from 10 Mbps to 10 Gbps speeds

41
Whats Next For Ethernet?

• 40 Gbps implementations are underway
• 100 Gbps could be possible by 2006
• Terabit (1000 Gigabit) may be seen by 2011 and 10
  Terabit by 2015
• Major implications for these tremendous rates of
  speed in the areas of entertainment and business

42
Ethernet Frame Types

• Four unique Ethernet frame types
• Ethernet 802.3 used by IPX/SPX on Novell NetWare
  2.x or 3.x networks
• Ethernet 802.2 used by IPX/SPX on Novell 3.12 and
  4.x networks default with Microsoft NWLink
• Ethernet SNAP used with EtherTalk and mainframes
• Ethernet II used by TCP/IP
• Types must match for two devices to communicate
• Packet size ranges from 64 to 1518 bytes

43
Ethernet 802.3

• Also called Ethernet raw
• Does not completely comply with 802.3
  specifications
• Used with Novell NetWare 2.x or 3.x
• Figure 7-3 shows frame

44
Ethernet 802.3 Frame
45
Segmentation

• Breaking network down into manageable pieces
• Uses switch or router between network segments
• Allows for more efficient network traffic
• See Figure 7-5

46
Switch Segments Network
47
Wireless EthernetIEEE 802.11b, a, and g

• Uses access point (AP) as center of star network
• Workstations have wireless NICs
• CSMA/CA access method with acknowledgement for
  every packet
• Handshaking before transmission prevents hidden
  node problem
• 802.11b standard specifies transmission rate of
  11 Mbps 802.11a and g specify 54 Mbps
• No fixed segment lengths, but maximum distance
  usually 300 feet with no obstructions

48
Token Ring

• Developed by IBM
• Provides fast reliable transport using
  twisted-pair cable
• Wired in physical star topology
• Functions as logical ring
• See Figure 7-6 and Simulation 7-2

49
Token Ring Physical Star Functions as Logical
Ring
50
Token Ring Function

• Uses token-passing channel access method
• Receives token from Nearest Active Upstream
  Neighbor (NAUN)
• Passes token to Nearest Active Downstream
  Neighbor (NADN)
• Provides equal access to all computers
• Uses larger packets, between 4000 and 17,800
  bytes with no collisions
• Originally operated at 4 Mbps, but newer version
  increased speed to 16 Mbps

51
Beaconing

• Technique automatically isolates faults
• First computer powered on network becomes active
  monitor managing beaconing process
• Other computers are standby monitors
• Active computer sends special packet to nearest
  downstream neighbor every 7 seconds
• Packet announces address of active monitor
• Network is intact if packet travels around
  network and returns to active monitor
• Figure 7-7 shows ability to reconfigure network
  to avoid problem area

52
Token Ring Reconfiguration to Avoid Break
53
Hardware Components

• Uses Multistation Access Unit (MAU or MSAU) or
  Smart Multistation Access Unit (SMAU)
• Two ports connect hubs in a ring
• Ring Out (RO) port on one hub connects to Ring In
  (RI) port on next hub to form ring
• IBMs implementation allows connection of 33 hubs
• Originally maximum of 260 stations per network
  now doubled to 520 maximum

54
Cabling in a Token Ring Environment

• IBM defined cable types
• Based on American Wire Gauge (AWG) standard that
  specified wire diameters
• See Table 7-8
• Table 7-9 summarizes token ring

55
IBM/Token Ring Cabling
56
Summary of Token Ring
57
AppleTalk and ARCnet

• Designed by Apple Computers, Inc., for Macintosh
  networks
• ARCnet rarely used today
• LocalTalk is physical implementation of AppleTalk

58
AppleTalk Environment

• Simple, easy-to-implement network architecture
• Uses built-in network interface on Macintoshes
• AppleTalk refers to overall network architecture,
  while LocalTalk refers to cabling system
• Uses dynamic addressing scheme
• Computer chooses numeric address and broadcasts
  it to make sure it is unused

59
AppleTalk Environment (continued)

• Phase 1 supported only 32 computers per network
  but was later increased to 254 computers and
  devices
• Phase 2 introduced EtherTalk and TokenTalk
• Allowed AppleTalk protocols to operate over
  Ethernet and token ring networks, respectively
• Increased maximum computers on AppleTalk network
  to more than 16 million

60
FDDI

• Fiber Distributed Data Interface
• Uses token-passing channel access method
• Features dual counter-rotating rings for
  redundancy, as seen in Figure 7-10
• Transmits at 100 Mbps
• Includes up to 500 nodes over distance of 100 km
  (60 miles)
• Wired as physical ring, uses no hubs
• Can use concentrators as central connection point

61
FDDI Network with Counter-Rotating Rings
62
FDDI (continued)

• Computer with token can send more than one data
  frame
• Avoids collisions by calculating network latency
• Can assign priority level to particular station
  or type of data
• Dual counter-rotating rings
• Data travels on primary ring
• In case of break, data moves to secondary ring,
  as shown in Figure 7-11

63
Dual Rings in FDDI Ensures Data Reaches
Destination
64
FDDI (continued)

• Uses two types of NICs
• Dual Attachment Stations (DAS) attaches to both
  rings used for servers and concentrators
• Single Attachment Stations (SAS) connects to
  only one ring used for workstations attached to
  concentrators
• Table 7-11 summarizes FDDI architecture

65
Summary of FDDI
66
Other Networking Alternatives

• Many broadband technologies, including
• Cable modem
• Digital Subscriber Line (DSL)
• Broadcast technologies
• Asynchronous Transfer Mode (ATM)

67
Broadband Technologies

• Use analog techniques to encode information
  across continuous range of values
• Baseband uses digital encoding scheme at single,
  fixed frequency
• Uses continuous electromagnetic or optical waves
• Two channels necessary to send and receive
• Offers extremely high-speed, reliable connectivity

68
Cable Modem Technology

• Delivers Internet access over standard cable
  television coaxial cable
• Official standard is Data-Over-Cable Service
  Interface Specification (DOCSIS)
• Uses asymmetrical communication with different
  downstream and upstream rates
• Upstream may be 10 Mbps
• Downstream usually between 256 Kbps and 1 Mbps
• See Figure 7-12

69
Typical Cable Modem Network
70
Digital Subscriber Line (DSL)

• Uses existing phone lines to carry voice and data
  simultaneously
• Most prominent variety is Asymmetric DSL (ADSL)
• Downloads and upload speeds differ significantly
• Download speeds from 256 Kbps to 8 Mbps
• Upload speeds from 16 Kbps to 640 Kbps
• Divides phone line into two frequency ranges,
  with frequencies below 4 KHz used for voice

71
Broadcast Technologies

• Provides Internet access by satellite television
  systems
• User connects to service provider by regular
  modem
• Service provider, such as DirectTV, sends data
  to satellite at speeds up to 400 Kbps

72
Asynchronous Transfer Mode (ATM)

• Designed for both LANs and WANs
• Uses connection-oriented switches and continuous
  dedicated circuit between two end systems
• Data travels in fixed short 53-byte cells with 5
  bytes for header and 48 bytes for data
• Enables guaranteed quality of service (QOS)
• Choice for long-haul high-bandwidth applications

73
ATM and SONET Signaling Rates

• ATM bandwidth rated in terms of optical carrier
  level in form OC-x
• X represents multiplier of basic OC-1 carrier
  rate of 51,840 Mbps
• Rate originally defined for Synchronous Optical
  Network (SONET)
• Table 7-12 lists common SONET optical carrier
  rates ranging from OC-1 to OC-3072
• Typical ATM rates range from OC-3 to OC-12

74
Optical Carrier Signaling Rates
75
High Performance Parallel Interface (HIPPI)

• Originally used with super-computers and high-end
  workstations
• Serial HIPPI is fiber-optic version
• Uses series of point-to-point optical links
• Provides bandwidth up to 800 Mbps
• Commonly used as network backbone prior to
  advent of Gigabit Ethernet
• HIPPI-6400, now known as Gigabyte System Network
  (GSN), transfers at 6.4 Gbps