Networks and Communication Lecture 3: Datalink WAN

1
Networks and CommunicationLecture 3 Datalink -
WAN

• Peter Steenkiste
• School of Computer Science
• Carnegie Mellon University
• ECOM, Summer 2000

2
Todays Lecture

• Encoding and framing.
• Datalink layer functions.
• Switching.
• Flow control and error control.

3
From Signals to Packets
4
Why Do We Need Encoding?

• Meet certain electrical constraints.
• Receiver needs enough transitions to keep track
  of the transmit clock
• Avoid receiver saturation
• Create control symbols, besides regular data
  symbols.
• E.g. start or end of frame, escape, ...
• Error detection or error corrections.
• Some codes are illegal so receiver can detect
  certain classes of errors
• Minor errors can be corrected by having multiple
  adjacent signals mapped to the same data symbol

5
Example Ethernet Manchester Encoding
0
1
1
0
.85
V
0
-.85
.1?s

• Positive transition for 0, negative for 1
• Transition every cycle communicates clock (but
  need 2 transition times per bit)
• DC balance has good electrical properties

6
4B/5B Encoding

• NRZI line code.
• Data coded as symbols of 5 line bits gt 4 data
  bits, so 100 Mbps uses 125 MHz.
• Uses less frequency space than Manchester
  encoding
• Each valid symbol has at least two 1s get dense
  transitions.
• 16 data symbols, 8 control symbols
• Data symbols 4 data bits
• Control symbols idle, begin frame, etc.
• Example FDDI.

7
Framing

• A link layer function, defining which bits have
  which function.
• Minimal functionality mark off units of
  transmission.
• Some techniques
• frame delimiter characters with character
  stuffing
• frame delimiter codes with bit stuffing
• out of band delimiters (e.g. FDDI control
  symbols)
• synchronous transmission (e.g. SONET)

8
Example Ethernet Framing
preamble
stuff
length
more stuff

• preamble is 7 bytes of 10101010 (5 MHz square
  wave) followed by one byte of 10101011
• allows receivers to recognize start of
  transmission after idle channel (an out of band
  signal)

9
Character and Bit Stuffing

• Mark frames with control characters, e.g. 4B/5B.
• Mark frames with special character.
• use escape character when controls appear in
  data
• abcdef -gt abc\def
• used all the time on serial lines, in editors,
  etc.
• Mark frames with special bit sequence
• must ensure data containing this sequence can be
  transmitted
• example suppose 11111111 is a special sequence.
• transmitter inserts a 0 when this appears in the
  data
• 11111111 -gt 111111101
• must stuff a zero any time seven 1s appear
• 11111110 -gt 111111100
• receiver destuffs.

10
SONET

• SONET is the Synchronous Optical Network
  standard for digital transport over optical
  fiber.
• One of the design goals was to be backwards
  compatible with many older telco standards.
• Beside minimal framing functionality, it provides
  many other functions
• operation, administration and maintenance (OAM)
  communications
• synchronization
• multiplexing of lower rate signals
• multiplexing for higher rates

11
SONET Framing

• Each STS-1 (Synchronous Transport System) frame
  takes 125 ?sec and contains 9 rows of 90 bytes
  each gt 51.84 Mbps. (OC-1 optical carrier.)
• 1 byte corresponds to a 64 Kbs channel (PCM
  voice)
• can hold a synchronous payload envelope (SPE)
• Transmission is by column within row.

3 cols transport overhead
87 cols payload capacity, including 1 col path
overhead
9 rows
12
The SONET Signal Hierarchy
ANSI SONET
ITU SDH
line rate
SPE rate
STS-1
51.84
50.112
STS-3
STM-1
155.52
150.336
STS-9
STM-3
466.56
451.008
STS-12
STM-4
622.08
601.344
STS-18
STM-6
933.12
902.016
STS-24
STM-8
1244.16
1202.688
STS-36
STM-12
1866.24
1804.032
STS-48
STM-16
2488.32
2405.376
STS-96
STM-32
4976.64
4810.752
STS-192
STM-64
9953.28
9621.504
OC Optical Carrier
13
Using SONET in Networks
Add-drop capability allows soft configuration of
networks, self-healing rings.
mux
OC-48
mux
DS1
OC-3c
OC-12c
mux
14
Error Coding

• Transmission process may introduce errors into a
  message.
• Single bit errors versus burst errors
• Detection
• Requires a convention that some messages are
  invalid
• Hence requires extra bits
• An (n,k) code has codewords of n bits with k data
  bits and r (n-k) redundant check bits
• Correction
• Forward error correction many related code words
  map to the same data word
• Detect errors and retry transmission

15
Basic ConceptHamming Distance

• Hamming distance of two bit strings number of
  bit positions in which they differ.
• If the valid words of a code have minimum Hamming
  distance D, then D-1 bit errors can be detected.
• If the valid words of a code have minimum Hamming
  distance D, then (D-1)/2 bit errors can be
  corrected.

1
0
1
1
0
HD2
1
1
0
1
0
HD3
16
Examples

• A (4,3) parity code has D2
• 0001 0010 0100 0111 1000 1011 1101 1110
• A (7,4) code with D3
• 0000000 0001101 0010111 0011010
• 0100011 0101110 0110100 0111001
• 1000110 1001011 1010001 1011100
• 1100101 1101000 1110010 1111111
• 1001111 corrects to 1001011
• Note the inherent risk in correction consider a
  2-bit error resulting in 1001011 -gt 1111011.
• There are formulas to calculate the number of
  extra bits that are needed to for a certain D.

17
Cyclic Redundancy Codes(CRC)

• Commonly used codes that have good error
  detection properties.
• Can catch many error combinations with a small
  number or redundant bits
• Based on division of polynomials.
• Errors can be viewed as adding terms to the
  polynomial
• Should be unlikely that the division will still
  work
• Can be implemented very efficiently in hardware.
• Examples
• CRC-32 Ethernet
• CRC-8, CRC-10, CRC-32 ATM

18
MAC/Logical Link Control

• How do we transfer packets between two hosts
  connected to the same network?
• Switches connected by point-to-point links --
  store-and-forward.
• Used in WAN, LAN, and for home connections
• Discussed today
• Multiple access networks -- broadcast based
• Multiple hosts are sharing the same transmission
  medium
• Used primarily in LANs and wireless
• Need to control access to the medium
• Contention based networks, e.g. Ethernet
• Token based, e.g. FFDI
• Reservation based, e.g. DQDB

19
A Switch-based Network

• Switches are connected by point-point links.
• Packets are forwarded hop-by-hop by the switches
  towards the destination.
• Forwarding is based on the address
• How do nodes exchange packets over a link?
• How does a switch work?
• How do adjacent switches manage the link?

Point-Point link
Switch
PCs at Work
PC at Home
20
Switch Architecture

• Takes in packets in one interface and has to
  forward them to an output interface based on the
  address.
• A big intersection
• Same idea for bridges, switches, routers address
  look up differs
• Control processor manages the switch and executes
  higher level protocols.
• E.g. routing, management, ..
• The switch fabric directs the traffic to the
  right output port.
• The input and output ports deal with transmission
  and reception of packets.

Control Processor
Switch Fabric
Input Port
Output Port
Output Port
Input Port
Output Port
Output Port
Input Port
Input Port
21
Input and Output Port Functions

• Input port identifies the outgoing port and
  buffers packets if there is contention for the
  switch fabric.
• Output port queues packets and a scheduler
  determines the order in which packets are sent
  over the outgoing link.
• Many buffering options exist.
• Input buffering, output buffering, internal
  buffering
• Typically a combination is used
• Buffer management can limit throughput, e.g. head
  of line blocking

Switch Fabric
Address Lookup
Scheduler
Address Lookup
22
Switch Fabric Options

• Crossbar switch.
• Requires lots of hardware but good performance
• Multistage interconnection networks an
  alternative
• Bus-based switches.
• Fabric consists one (or more) fast shared buses
• Each input port has a slot time slot on the bus
• Shared memory switch.
• Switch is one large memory
• Input ports write packets to memory and output
  ports read packets from memory
• Does not scale well need very fast memory
• Hybrid solutions.

23
A Simple Bus-based Switch
Input Ports
Bus
Output Ports
24
A Crossbar Architecture
Input Ports
Output Ports
25
The Knockout Switch
input buses
concentrators
buffers
26
How are Switches Packaged?

• Low-end switches are typically sold as a closed
  system.
• Commodity part - no options
• Mid-range switches typically offer some options.
• E.g. optional high speed interface, ATM or
  Gigabit Ethernet
• High-end switches are assembled on spec.
• Chassis has the switch fabric and empty slots
• Control card and power supply can be part of the
  chassis or separate
• Customer specifies the desired interface cards
• Interface cards are available for both high-speed
  (single interface/card) to low-speed links (many
  interfaces/card)

27
Performance Features to Look For

• Throughput of the backplane in Mbit/second.
• Limits the throughput of all links combined
• Throughput in packets/second
• Limit is often on a per-interface basis
• Typically limits the throughput for small packets
• Redundant power supplies and/or control
  processor.
• Switch will continue to work after certain
  failures
• Hot swappable can change cards without bringing
  down the switch.
• Configuration can be changed or bad interface
  cards can be replaced without affecting the other
  links

28
Link Flow Control and Error Control

• Naïve protocol.
• Dealing with receiver overflow flow control.
• Dealing with packet loss and corruption error
  control.
• Meta-comment these issues are relevant at many
  layers.
• Link layer sender and receiver attached to the
  same wire
• End-to-end transmission control protocol (TCP) -
  sender and receiver are the end points of a
  connection

29
A Naïve Protocol

• Sender simply sends to the receiver whenever it
  has packets.
• Potential problem sender can outrun the
  receiver.
• Receiver too slow, buffer overflow, ..
• Not always a problem receiver might be fast
  enough.

Receiver
Sender
30
Adding Flow Control

• Stop and go flow control sender waits to send
  the next packet until the previous packet has
  been acknowledged by the receiver.
• Receiver can pace the receiver
• Drawbacks adds overheads, slowdown for long
  links.

Receiver
Sender
31
Window Flow Control

• Stop and go flow control results in poor
  throughput for long-delay paths packet size/
  roundtrip-time.
• Solution receiver provides sender sender with a
  window that it can fill with packets.
• The window is backed up by buffer space on
  receiver
• Receiver acknowledges the a packet every time a
  packet is consumed and a buffer is freed

Receiver
Sender
32
The Bandwidth-Delay Product of a Link
Sender
Receiver
Time
Window Size
Throughput
Roundtrip Time
33
A Danger of Flow ControlBack pressure through
the Network

• Back pressure propagates upstream and creates a
  cone of congestion.
• Can be avoided using specific scheduling
  techniques
• The alternative is to drop the packets -
  sometimes a good idea!

34
Dealing with Errors

• Packet corruption.
• Use error detection or correction - results in
  lost or correct packet
• Receiver can ask for new copy or consider the
  packet lost
• Assume a stop and go protocol
• Lost packets add time out mechanisms.
• Positive versus negative acknowledgements
• Need sequence numbers in the case of window flow
  control

Receiver
Sender
35
What is Used in Practice?

• No flow or error control.
• E.g. regular Ethernet, just uses CRC for error
  detection
• Flow control only.
• E.g. Gigabit Ethernet
• Flow and error control.
• E.g. X.25 (older connection-based service at 64
  Kbs that guarantees reliable in order delivery of
  data)

36
Wide-Area Connections

• X.25, T1, T3, Packets over SONET.
• Standards for transfer of packets over different
  media
• Trend older technologies include more error
  control, while newer technologies use simpler
  protocols
• ATM Asynchronous Transfer Mode.
• Hot 8 years ago cooler now
• Wireless connections.