Week 3 Virtual LANs, Wireless LANs, PPP, ATM

1
Week 3Virtual LANs, Wireless LANs, PPP, ATM

2
Virtual LANs

• It is the territory over which a broadcast or
  multicast packet is delievered (also known as a
  broadcast domain)
• The difference in a VLAN and a LAN, if there is
  any, is in packaging
• Virtual LANs allow you to have separate LANs
  among ports on the same switch
• For example, a switch might be told that ports
  1-32 are in VLAN A and ports 33-64 are in VLAN B

3
Virtual and Physical LANs
4
Why VLANs

• IP requires that all nodes on a LAN share the
  same IP address prefix therefore a node that
  moves to a different LAN must change its address
• Changing IP addresses manually is annoying
• IP broadcasts traffic within a LAN, something
  that can cause congestion in a large LAN
• Routing IP (rather than bridging) was slow
• It might be tempting to bridge everything making
  your whole topology one giant LAN from the
  perspective of IP and use layer 2 switches

5
Disadvantages of one single LAN

• Broadcast traffic (such as ARP) grows in
  proportion to the number of stations
• Users can snoop on the traffic of other users on
  the same LAN, so it might be safer to isolate
  groups of users onto different LANs
• Some protocols are overly chatty or they get into
  modes such as broadcast storms.
• So it seems desirable for users that need to talk
  to each other a lot to be in the same LAN but
  keep other groups of users in separate LANs
• A VLAN makes us broadcast domain as large as we
  want it

6
Mapping ports to VLANs

• The switch has ports 1 to k in one VLAN and has
  ports k1 to 2k in another LAN
• The switch can be configured with a port/VLAN
  mapping
• The switch can be configured with a table of
  VLAN/MAC address mappings. It then dynamically
  determines the VLAN/port mapping based on the
  learned MAC address of the station attached to
  the port.
• The switch can be configured with a table of
  VLAN/IP prefix mappings. It then dynamically
  determines the VLAN/port mapping based on the
  source IP address from the station attached to
  the port.
• The switch can be configured with a table of
  VLAN/protocol mappings. It then dynamically
  determines the VLAN/port mappings based on the
  protocol type of the stations attached to the
  port.

7
VLAN forwarding with separate router
a.b.c.H
f.g.k.Q
h
q
2
9
d
3
x
13
a.b.c.D
11
7
f.g.k.X
j
f
a.b.c.R1
f.g.k.R2
Router R
Router connects VLANs
8
VLAN forwarding with switch as router
a..b.c.H
f.g.k.Q
9
2
Switch/ Router R
VLAN A
VLAN B
3
12
a..b.c.D
f.g.k.X

• Router does not use up ports
• The switch must know that Rs mac address on VLAN
  A is f and on VLAN B is j.

9
Dynamic binding of links to VLANs
The switch now learns that there are two VLANs on
port a If enough stations move around, advantage
disappears
10
VLAN Tagging
VLAN 2
VLAN 1
VLAN 2
VLAN 2
VLAN1
VLAN 1
Interswitch port Packets can belong in
either VLAN1 or VLAN2 IEEE standardized a scheme
for VLAN tagging
VLAN 1
VLAN 1
VLAN 2
11
IEEE 802.11 Wireless LAN

• 802.11b
• 2.4-5 GHz unlicensed radio spectrum
• up to 11 Mbps
• direct sequence spread spectrum (DSSS) in
  physical layer
• all hosts use same chipping code
• widely deployed, using base stations

• 802.11a
• 5-6 GHz range
• up to 54 Mbps
• 802.11g
• 2.4-5 GHz range
• up to 54 Mbps
• All use CSMA/CA for multiple access
• All have base-station and ad-hoc network versions

12
802.11 LAN architecture

• wireless host communicates with base station
• base station access point (AP)
• Basic Service Set (BSS) (aka cell) in
  infrastructure mode contains
• wireless hosts
• access point (AP) base station
• ad hoc mode hosts only

hub, switch or router
BSS 1
BSS 2
13
802.11 Channels, association

• 802.11b 2.4GHz-2.485GHz spectrum divided into 11
  channels at different frequencies
• AP admin chooses frequency for AP
• interference possible channel can be same as
  that chosen by neighboring AP!
• host must associate with an AP
• scans channels, listening for beacon frames
  containing APs name (SSID) and MAC address
• selects AP to associate with
• may perform authentication
• will typically run DHCP to get IP address in APs
  subnet

14
IEEE 802.11 multiple access

• avoid collisions 2 nodes transmitting at same
  time
• 802.11 CSMA - sense before transmitting
• dont collide with ongoing transmission by other
  node
• 802.11 no collision detection!
• difficult to receive (sense collisions) when
  transmitting due to weak received signals
  (fading)
• cant sense all collisions in any case hidden
  terminal, fading
• goal avoid collisions CSMA/C(ollision)A(voidance
  )

15
IEEE 802.11 MAC Protocol CSMA/CA

• 802.11 sender
• 1 if INITIALLY sense channel idle for DIFS then
• transmit entire frame (no CD)
• 2 if sense channel busy then
• start random backoff time
• timer counts down while channel idle
• transmit when timer expires
• if no ACK, increase random backoff interval,
  repeat 2
• 802.11 receiver
• - if frame received OK
• return ACK after SIFS (ACK needed due to
  hidden terminal problem)

sender
receiver
16
Avoiding collisions (more)

• idea allow sender to reserve channel rather
  than random access of data frames avoid
  collisions of long data frames
• sender first transmits small request-to-send
  (RTS) packets to BS using CSMA
• RTSs may still collide with each other (but
  theyre short)
• BS broadcasts clear-to-send CTS in response to
  RTS
• RTS heard by all nodes
• sender transmits data frame
• other stations defer transmissions

Avoid data frame collisions completely using
small reservation packets!
17
Collision Avoidance RTS-CTS exchange
A
B
AP
defer
time
18
802.11 frame addressing
Address 4 used only in ad hoc mode
Address 1 MAC address of wireless host or AP to
receive this frame
Address 3 MAC address of router interface to
which AP is attached
Address 2 MAC address of wireless host or AP
transmitting this frame
19
802.11 frame addressing
H1
R1
20
802.11 frame more
frame seq (for reliable ARQ)
duration of reserved transmission time (RTS/CTS)
frame type (RTS, CTS, ACK, data)
21
802.11 mobility within same subnet

• H1 remains in same IP subnet IP address can
  remain same
• switch which AP is associated with H1?
• self-learning (Ch. 5) switch will see frame from
  H1 and remember which switch port can be used
  to reach H1

hub or switch
BBS 1
AP 1
AP 2
H1
BBS 2
22
Point to Point Data Link Control

• one sender, one receiver, one link easier than
  broadcast link
• no Media Access Control
• no need for explicit MAC addressing
• e.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!

23
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 carry any bit pattern in
  the data field
• error detection (no correction)
• connection liveness detect, signal link failure
  to network layer
• network layer address negotiation endpoint can
  learn/configure each others network address

24
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 higher layers!
25
PPP Data Frame

• 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, IP, IPCP, etc)

26
PPP Data Frame

• info upper layer data being carried
• check cyclic redundancy check for error
  detection

27
Byte Stuffing

• data transparency requirement data field must
  be allowed to include flag pattern lt01111110gt
• Q is received lt01111110gt data or flag?
• Sender adds (stuffs) extra lt 01111101gt byte
  before each lt 01111110gt data byte
• Receiver
• 01111101 and 01111110 bytes in a row discard
  first byte, continue data reception
• single 01111110 flag byte

28
Byte Stuffing
flag byte pattern in data to send
flag byte pattern plus stuffed byte in
transmitted data
29
PPP Data Control Protocol

• Before exchanging network-layer data, data link
  peers must
• configure PPP link (max. frame length,
  authentication)
• learn/configure network
• layer information
• for IP carry IP Control Protocol (IPCP) msgs
  (protocol field 8021) to configure/learn IP
  address

30
Link Layer

• 5.1 Introduction and services
• 5.2 Error detection and correction
• 5.3Multiple access protocols
• 5.4 Link-Layer Addressing
• 5.5 Ethernet

• 5.6 Hubs and switches
• 5.7 PPP
• 5.8 Link Virtualization ATM and MPLS

31
Virtualization of networks

• Virtualization of resources a powerful
  abstraction in systems engineering
• computing examples virtual memory, virtual
  devices
• Virtual machines e.g., java
• IBM VM os from 1960s/70s
• layering of abstractions dont sweat the details
  of the lower layer, only deal with lower layers
  abstractly

32
The Internet virtualizing networks

• 1974 multiple unconnected nets
• ARPAnet
• data-over-cable networks
• packet satellite network (Aloha)
• packet radio network

• differing in
• addressing conventions
• packet formats
• error recovery
• routing

satellite net
ARPAnet
"A Protocol for Packet Network Intercommunication"
, V. Cerf, R. Kahn, IEEE Transactions on
Communications, May, 1974, pp. 637-648.
33
The Internet virtualizing networks

• Gateway
• embed internetwork packets in local packet
  format or extract them
• route (at internetwork level) to next gateway

gateway
satellite net
ARPAnet
34
Cerf Kahns Internetwork Architecture

• What is virtualized?
• two layers of addressing internetwork and local
  network
• new layer (IP) makes everything homogeneous at
  internetwork layer
• underlying local network technology
• cable
• satellite
• 56K telephone modem
• today ATM, MPLS
• invisible at internetwork layer. Looks
  like a link layer technology to IP!

35
Generic connection oriented network

• For A to talk to B, there must be a special call
  setup packet that travels from A to B, specifying
  B as the destination.
• Each router along the path must make a routing
  decision based on Bs address
• This is the identical problem in IP
• In addition to simply forwarding the call setup
  packet, the goal is to assign the call a small
  identifier, which we now call the CI (connection
  identifier)
• CIs can be small because they are handed out
  dynamically and are significant only on a link
• They only need to be large enough to distinguish
  between the total number of calls that might
  simultaneously be routed on the same link

36
A wants to talk to B and use CI 57
22?b,79
79?c,22
57? c,33
c
33?d,79
33 ? a,57
b
79?a,33
a
c
c
a
a
b

• Why does the CI have to change hop by hop?
• The answer is that it would be very difficult to
  choose a CI that was unused on all the links
  along the path

37
ATM and MPLS

• ATM, MPLS separate networks in their own right
• different service models, addressing, routing
  from Internet
• viewed by Internet as logical link connecting IP
  routers
• just like dialup link is really part of separate
  network (telephone network)
• ATM, MPLS of technical interest in their own
  right

38
Asynchronous Transfer Mode ATM

• 1990s/00 standard for high-speed (155Mbps to 622
  Mbps and higher) Broadband Integrated Service
  Digital Network architecture
• Goal integrated, end-end transport of carry
  voice, video, data
• meeting timing/QoS requirements of voice, video
  (versus Internet best-effort model)
• next generation telephony technical roots in
  telephone world
• packet-switching (fixed length packets, called
  cells) using virtual circuits

39
ATM architecture

• adaptation layer only at edge of ATM network
• data segmentation/reassembly
• roughly analagous to Internet transport layer
• ATM layer network layer
• cell switching, routing
• physical layer

40
ATM network or link layer?

• Vision end-to-end transport ATM from desktop
  to desktop
• ATM is a network technology
• Reality used to connect IP backbone routers
• IP over ATM
• ATM as switched link layer, connecting IP routers

IP network
ATM network
41
ATM Adaptation Layer (AAL)

• ATM Adaptation Layer (AAL) adapts upper layers
  (IP or native ATM applications) to ATM layer
  below
• AAL present only in end systems, not in switches
• AAL layer segment (header/trailer fields, data)
  fragmented across multiple ATM cells
• analogy TCP segment in many IP packets

42
ATM Adaptation Layer (AAL) more

• Different versions of AAL layers, depending on
  ATM service class
• AAL1 for CBR (Constant Bit Rate) services, e.g.
  circuit emulation
• AAL2 for VBR (Variable Bit Rate) services, e.g.,
  MPEG video
• AAL5 for data (eg, IP datagrams)

User data
AAL PDU
ATM cell
43
ATM Layer

• Service transport cells across ATM network
• analogous to IP network layer
• very different services than IP network layer

Guarantees ?
Network Architecture Internet ATM ATM ATM ATM
Service Model best effort CBR VBR ABR UBR
Congestion feedback no (inferred via
loss) no congestion no congestion yes no
Bandwidth none constant rate guaranteed rate gua
ranteed minimum none
Loss no yes yes no no
Order no yes yes yes yes
Timing no yes yes no no
44
ATM Layer Virtual Circuits

• VC transport cells carried on VC from source to
  dest
• call setup, teardown for each call before data
  can flow
• each packet carries VC identifier (not
  destination ID)
• every switch on source-dest path maintain state
  for each passing connection
• link,switch resources (bandwidth, buffers) may be
  allocated to VC to get circuit-like perf.
• Permanent VCs (PVCs)
• long lasting connections
• typically permanent route between to IP
  routers
• Switched VCs (SVC)
• dynamically set up on per-call basis

45
ATM VCs

• Advantages of ATM VC approach
• QoS performance guarantee for connection mapped
  to VC (bandwidth, delay, delay jitter)
• Drawbacks of ATM VC approach
• Inefficient support of datagram traffic
• one PVC between each source/dest pair) does not
  scale (N2 connections needed)
• SVC introduces call setup latency, processing
  overhead for short lived connections

46
ATM Layer ATM cell

• 5-byte ATM cell header
• 48-byte payload
• Why? small payload -gt short cell-creation delay
  for digitized voice
• halfway between 32 and 64 (compromise!)

Cell header
Cell format
47
ATM cell header

• VCI virtual channel ID
• will change from link to link thru net
• PT Payload type (e.g. RM cell versus data cell)
• CLP Cell Loss Priority bit
• CLP 1 implies low priority cell, can be
  discarded if congestion
• HEC Header Error Checksum
• cyclic redundancy check

48
ATM VCs

• Advantages of ATM VC approach
• QoS performance guarantee for connection mapped
  to VC (bandwidth, delay, delay jitter)
• Drawbacks of ATM VC approach
• Inefficient support of datagram traffic
• One PVC between each source/dest pair) does not
  scale (N2 connections needed)
• SVC introduces call setup latency, processing
  overhead for short lived connections

49
Virtual Path Concept

• The connection identifier in the ATM cell header
  has two complexities
• Its hierarchical and divided into two subfields
  VPI (Virtual Path Identifier) and VCI (Virtual
  Circuit Identifier)
• VCI is 16 bits
• VPI is 12 bits
• Whats a VPI? There might be very high speed
  backbone carrying many millions of calls
• The split between VPI and VCI saves the switches
  in the backbone from requiring that their call
  mapping database keep track of millions of calls

50
Virtual Path Concept

• The backbone routers only use the VPIU field then
  if needed
• Outside the backbone, the switches treat the
  entire VPIVCI field as one nonhierarchical unit
• VP switch looks at only the VPI portion
• VC switch looks at both

51
Example
S2
D
S3
a
b
S1
e
c
S4
S5

• S1 is to receive a call setup on port b with CI
  17 for destination D
• Normal VP switching inside the core with the CI
  being the
• 12 bit VPI
• Switches outside the core do normal VC
  switching with the CI
• being 28 bits
• Switches at the border also do VC swiching but
  the outgoing CI
• must be chosen so that the VPI portion of the
  outgoing CI is
• to the outgoing VPI

52
Example
89?c,187.42
187. ? d,13
13. ? e,57
57.42? d,83
64000 VCs can be carried within a single VP
dramatically reducing the switch table sizes
53
Virtual Path and Virtual Channels
Virtual Channels (VC)
ATM Physical LinkVirtual Channel Connection (VCC)
Virtual Path (VP)
E3OC12
Virtual Path (VP)
Virtual Channels (VC)
Virtual Channel(VC)Logical PathBetween ATM End
Points
Virtual Path(VP)Contains Multiple VCs
Virtual Channel Connection(VCC)Contains
Multiple VPs
Connection Identifier VPI/VCI
54
ATM Switches
Input
Output
45
VPI/VCI
Port
VPI/VCI
Port
29
1
45
2
64
29
45
2
29
1
64
1
29
3
29
3
64
1
29

• ATM switches translate VPI/VCI values
• VPI/VCI value unique only per interfaceeg
  locally significant and may be re-used elsewhere
  in network

55
VP and VC Switching
VC Switch
VCI 1
VCI 2
VCI 3
VCI 4
Port 2
VPI 2
VPI 3
VPI 1
VP Switch
VPI 2
Port 1
VCI 1
VPI 3
VPI 1
VCI 2
VCI 1
VPI 5
VPI 4
VCI 2
Port 3
56
Virtual Channels and Virtual Paths
Virtual Channel Connection (VCC)
Virtual PathConnection (VPC)
UNI
UNI
NNI
NNI
VPSwitch
VCSwitch
VCSwitch
VPI 2VCI 44
VPI 1VCI 1
VPI 26VCI 44
VPI 20VCI 30

• This hop-by-hop forwarding is known as cell relay

57
Virtual Path and Virtual Channels
Virtual Channels (VC)
ATM Physical LinkVirtual Channel Connection (VCC)
Virtual Path (VP)
E3OC12
Virtual Path (VP)
Virtual Channels (VC)
Virtual Channel(VC)Logical PathBetween ATM End
Points
Virtual Path(VP)Contains Multiple VCs
Virtual Channel Connection(VCC)Contains
Multiple VPs
Connection Identifier VPI/VCI
58
ATM Switches
Input
Output
45
VPI/VCI
Port
VPI/VCI
Port
29
1
45
2
64
29
45
2
29
1
64
1
29
3
29
3
64
1
29

• ATM switches translate VPI/VCI values
• VPI/VCI value unique only per interfaceeg
  locally significant and may be re-used elsewhere
  in network

59
VP and VC Switching
VC Switch
VCI 1
VCI 2
VCI 3
VCI 4
Port 2
VPI 2
VPI 3
VPI 1
VP Switch
VPI 2
Port 1
VCI 1
VPI 3
VPI 1
VCI 2
VCI 1
VPI 5
VPI 4
VCI 2
Port 3
60
Virtual Channels and Virtual Paths
Virtual Channel Connection (VCC)
Virtual PathConnection (VPC)
UNI
UNI
NNI
NNI
VPSwitch
VCSwitch
VCSwitch
VPI 2VCI 44
VPI 1VCI 1
VPI 26VCI 44
VPI 20VCI 30

• This hop-by-hop forwarding is known as cell relay

61
Example
62
ATM Physical Layer (more)

• Two pieces (sublayers) of physical layer
• Transmission Convergence Sublayer (TCS) adapts
  ATM layer above to PMD sublayer below
• Physical Medium Dependent depends on physical
  medium being used
• TCS Functions
• Header checksum generation 8 bits CRC
• Cell delineation
• With unstructured PMD sublayer, transmission of
  idle cells when no data cells to send

63
ATM Physical Layer

• Physical Medium Dependent (PMD) sublayer
• SONET/SDH transmission frame structure (like a
  container carrying bits)
• bit synchronization
• bandwidth partitions (TDM)
• several speeds OC3 155.52 Mbps OC12 622.08
  Mbps OC48 2.45 Gbps, OC192 9.6 Gbps
• TI/T3 transmission frame structure (old
  telephone hierarchy) 1.5 Mbps/ 45 Mbps
• unstructured just cells (busy/idle)

64
IP-Over-ATM

• IP over ATM
• replace network (e.g., LAN segment) with ATM
  network
• ATM addresses, IP addresses

• Classic IP only
• 3 networks (e.g., LAN segments)
• MAC (802.3) and IP addresses

ATM network
Ethernet LANs
Ethernet LANs
65
IP-Over-ATM
66
Datagram Journey in IP-over-ATM Network

• at Source Host
• IP layer maps between IP, ATM dest address (using
  ARP)
• passes datagram to AAL5
• AAL5 encapsulates data, segments cells, passes to
  ATM layer
• ATM network moves cell along VC to destination
• at Destination Host
• AAL5 reassembles cells into original datagram
• if CRC OK, datagram is passed to IP

67
IP-Over-ATM

• Issues
• IP datagrams into ATM AAL5 PDUs
• from IP addresses to ATM addresses
• just like IP addresses to 802.3 MAC addresses!

ATM network
Ethernet LANs
68
ATM Layer

• Service transport cells across ATM network
• analogous to IP network layer
• very different services than IP network layer

Guarantees ?
Congestion feedback no (inferred via
loss) no congestion no congestion yes no
Service Model best effort CBR VBR ABR UBR
Network Architecture Internet ATM ATM ATM ATM
Bandwidth none constant rate guaranteed rate gua
ranteed minimum none
Loss no yes yes no no
Order no yes yes yes yes
Timing no yes yes no no
69
Traffic Management

• Why traffic management?
• Traffic control techniques
• AAL5/ABR congestion feedback
• Buffers are your friend

70
Why Traffic Management?

• Proactively combat congestion
• Provision for priority control
• Maintain well-behaved traffic

71
Why Traffic Management?
Cell LossDatas Critical Enemy
Ethernet (1500 Bytes) 32 Cells FDDI (4470
Bytes) 96 Cells IP over ATM1577 (9180 Bytes)
192 Cells
TCP/IP Packet
X

• Lose one cell and the rest are useless
• Need to re-transmit 32 cells for one cell lost
• Congestion collapse is the result
• PPD (Partial Packet Discard)
• EPD (Early Packet Discard)

72
Traffic Control Techniques

• Connection managementAcceptance
• Traffic managementPolicing
• Traffic smoothingShaping

73
Traffic Control Techniques
Connection Management
Contract
ATM Network
Contract

• Traffic Parameters
• Peak cell rate
• Sustainable cell rate
• Burst tolerance
• Etc.
• Quality of Service
• Delay
• Cell loss

74
Traffic Descriptors

• Peak Cell Rate(PCR) 1/T in units of
  cells/second, where T is the minimum intercell
  spacing in seconds(i.e., the time interval from
  the first bit of one cell to the first bit of the
  next cell)
• Sustainable Cell Rate(SCR) is the maximum average
  rate that a bursty, on-off traffic source can be
  sent at the peak rate
• Maximum Burst Size(MBS) is the maximum number of
  cells that can be sent at the peak rate
  

75
QoS Expectations

• Applications have service requirements on
• Throughput
• Maximum Delay
• Variance of Delays(Delay Jitter)
• Loss Probability
• Network has to guarantee the required Quality of
  Service(Traffic Contract)
• Major Problem Bursty Traffic,
• i.e., Peak Traffic Rate gtgt Average Traffic
  Rate

76
Traffic Control Techniques
Connection Management Connection Admission
Control (CAC)
I want a VC X Mbps Y Delay Z Cell Loss
CAC Can I Support this Reliably without
Jeopardizing Other Contracts
Guaranteed QoS Request
No or Yes, Agree to aTraffic Contract
Contract
ATM Network
77
Connection Admission Control

• The primary function of the CAC is to accept a
  new connection request only if its stated QoS
  can be maintained without influencing the QoS of
  the already accepted connections.
• It is very likely that certain calls will
  require more than one connection (e.g.,
  teleconferencing) CAC procedure must be performed
  for each requested VCC or VPC.
• CAC must
• Decide whether connections can be accepted or
  not.
• Provide parameters required by the UPC.
• Perform routing and resource allocation.

78
Bandwidth Allocation

• Peak Allocation
• Suppose a source has an average BW of 20 Mbps and
  a peak BW of 45 Mbps. Peak BW allocation requires
  that 45 Mbps be reserved at the output port for
  the specific source independent of whether or not
  the source transmits continuously at 45 Mbps.
• Peak BW allocation is used for CBR services.
  The advantage of peak BW allocation is that it is
  easy to decide whether to accept a new connection
  or not.
• The new connection is accepted, if the sum of
  the peak rates of all the existing connections
  plus the peak rate of the new connection is less
  than the capacity of the output link.
• The disadvantage of the Peak BW allocation is
  that the output port link will be underutilized
  if the sources do not transmit at their peak
  rates.

79
Bandwidth Allocation

• Statistical Allocation
• The allocated BW is less than the peak rate of
  the source.
• The sum of all peak rates may be greater than the
  capacity of the output link.
• An equivalent capacity is allocated between the
  peak rate and the mean rate
• Call admission if the sum of the equivalent
  capacities is less than the capacity, reject the
  incoming call

80
Source Behavior
Cell Interarrival Time
CBR
time
Cell Interarrival Time
Burst Duration
Burst to Burst Interval
VBR
time
Call Duration
Call Tear-Down
Call Set-up
VBR Source Description
ON
OFF
Peak Arrival Rate Average Arrival Rate Mean Burst
Length
or
Burst Length Distribution Interarrival
Distribution During Burst Idle (silent) Length
Distribution
lt Bp, Bm, T gt
81
End-to-end Model
CAC is based on an abstract performance model of
the network.
Demultiplexing
Entering Cross Traffic
Multiplexing
Entering Cross Traffic
Entering Cross Traffic
Departing Cross Traffic
Departing Cross Traffic
Departing Cross Traffic
- FINITE BUFFERS - DETERMINISTIC SERVICE TIMES
Modeling Problems
Challenges - Arrival streams are non-Poisson -
Finite buffers at the multiplexers and switches -
Correlated cell arrivals - Large state-space of
the resulting system - Simulations of such
systems take very long to converge
82
Traffic Control Techniques
Traffic ManagementUsage Parameter Control (UPC)
aka Policing
You are Not in Conformance with the
Contract. What Should the Penalty Be??
Contract
?DECISION?
REBEL APPLICATION

• PASS
• MARK CLP BIT
• DROP

83
Traffic Control Techniques
Traffic Management
UPC
Marked
0
0
0
0
1
0
?DECISION?
D r o p

• PASS
• MARK CLP BIT
• DROP

• CLP ControlWhen congested drop marked cells
• Public UNIGeneric Cell Rate Algorithm (GCRA)

84
Policing

• The operation of the CAC and the correct
  allocation of resources depend heavily on the
  guarantee that the traffic source will behave as
  expected, i.e., as described by the traffic
  descriptor.
• Thus a monitoring/policing function is needed to
  force the traffic to comply to the traffic
  descriptor.
• This monitoring/policing function is performed
  by the UPC (policer).
• The UPC is in the form of preventive congestion
  control.
• It enforces a certain cell arrival rate or
  shape, such that it does not exceed certain
  values that would cause network elements to
  overload and lead to congestion.
• A UPC usually consists of a counter-based
  mechanism that drops or marks data units when
  they are found in violation of a certain
  agreement between end-user and the communication
  system.
• It does not use information from remote network
  elements.

85
Generic Cell Rate Algorithm (I, L)
The GCRA is reference algorithm for a cell rate
which determines if a cell is conforming.
Arrival of a cell k at time ta (k)
YES
TAT ? ta(k)
XX-(ta(k)-LCT)
YES
Xlt 0
TAT ta(k)
Non Conforming Cel
YES
Non Conforming Cell
YES
TAT lt ta(k) L
X0
Xgt L
NO
XXI LCT ta(k) Conforming Cell
TAT TAT I Conforming Cell
CONTINUOUS-STATE LEAKY BUCKET ALGORITHM
VIRTUAL SCHEULING ALGORITHM
X Value of the Leaky Bucket counter X
auxiliary variable LCT Last Compliance Time
I Increment L Limit
TAT Theoretical Arrival Time ta(k) Time of
arrival of a cell
Virtual Scheduling Algorithm TAT ta(1) initially
Leaky Bucket Algorithm X 0 LCT ta(1)
initially
86
Traffic Contact and Performance Definitions

• CBR
• GCRA(T01 , CDVT) in relation to the PCR01
• T01 is the inverse of PCR01
• Nonconformant cells are dropped
• VBR (one of the standardized definitions)
• GCRA(T01 , CDVT) in relation to the PCR01
• GCRA(Ts0 , BT0 CDVT) in relation to the SCR of
  the CLP 0 cell stream
• BT (MBS 1) (1/SCR - 1/PCR)
• If CLP 0 cell conforms to (1) and (2), that
  cell is conformant
• If CLP 0 cell is not conforming to (2) but is
  conforming to (1) then it will be remarked as CLP
  1

87
Example
88
Traffic Control Techniques
Traffic Management
UPC
Marked
0
0
0
0
1
0
D r o p
3
2

• Intelligent Packet DiscardIPD
• Discard cells from same bad packet
• Tail packet discard
• Maximize Goodput

89
Traffic Control Techniques
Traffic Smoothing
I Want to Comply With My Contract. So, I Will
Smooth/Shape My Traffic
Go Ahead, Make My Day
Shaper
Actual Data
Shaped Data
Private ATM Network
Public ATM Network

• Traffic shaper at customer site
• Changes traffic characteristics
• Leaky bucket algorithm

90
Traffic Control Techniques
Buffers Are Your Friend

• Absorb traffic bursts from simultaneous
  connections
• Switches schedule traffic based on priority of
  traffic according to QoS
• Switch must reallocate buffers as the traffic mix
  changes
• Effective buffering maximizes throughput of
  usable cells as opposed to raw cells (aka
  goodput)