Summary of Last Lecture Wireless

1
Summary of Last Lecture Wireless

• Elements of wireless networks
• 802.11 frequency band
• CDMA
• MAC transmission protocol
• CSMA/CA
• Hidden terminal problem
• Exposed terminal problem

2
Summary of Last Lecture

• Collision avoidance
• RTS/CTS exchange
• 802.11 frame
• 802.15 personal area network
• Bluetooth

3
Summary of Last LectureQueue management

• Congestion control
• Single link
• Leaky bucket
• Token bucket
• Multiple links
• Round robin
• Fair queuing
• Weighted fair queuing

4
Queuing theory
5
Queuing theory definitions

• (Bose) the basic phenomenon of queueing arises
  whenever a shared facility needs to be accessed
  for service by a large number of jobs or
  customers.
• (Wolff) The primary tool for studying these
  problems of congestions is known as queueing
  theory.
• (Kleinrock) We study the phenomena of standing,
  waiting, and serving, and we call this study
  Queueing Theory." "Any system in which arrivals
  place demands upon a finite capacity resource may
  be termed a queueing system.
• (Mathworld) The study of the waiting times,
  lengths, and other properties of queues.

http//www2.uwindsor.ca/hlynka/queue.html
6
Applications of Queuing Theory

• Telecommunications
• Traffic control
• Determining the sequence of computer operations
• Predicting computer performance
• Health services (eg. control of hospital bed
  assignments)
• Airport traffic, airline ticket sales
• Layout of manufacturing systems.

http//www2.uwindsor.ca/hlynka/queue.html
7
Example application of queuing theory

• In many retail stores and banks
• multiple line/multiple checkout system ? a
  queuing system where customers wait for the next
  available cashier
• We can prove using queuing theory that
  throughput improves when queues are used instead
  of separate lines

http//www.andrews.edu/calkins/math/webtexts/prod
10.htmQT
8
Example application of queuing theory
http//www.bsbpa.umkc.edu/classes/ashley/Chaptr14/
sld006.htm
9
Queuing theory for studying networks

• View network as collections of queues
• FIFO data-structures
• Queuing theory provides probabilistic analysis of
  these queues
• Examples
• Average length
• Average waiting time
• Probability queue is at a certain length
• Probability a packet will be lost

10
Littles Law
System
Arrivals
Departures

• Littles Law Mean number tasks in system mean
  arrival rate x mean response time
• Observed before, Little was first to prove
• Applies to any system in equilibrium, as long as
  nothing in black box is creating or destroying
  tasks

11
Proving Littles Law
Arrivals
Packet
Departures
1 2 3 4 5 6 7 8
Time
J Shaded area 9 Same in all cases!
12
Definitions

• J Area from previous slide
• N Number of jobs (packets)
• T Total time
• l Average arrival rate
• N/T
• W Average time a job is spent in the system
• J/N
• L Average number of jobs in the system
• J/T

13
Proof Method 1 Definition
in System (L)

1 2 3 4 5 6 7 8
Time (T)
14
Proof Method 2 Substitution
Tautology
15
Formalization Estimating Quantities

• The average number of jobs in the system
• Waiting in queue
• Undergoing service
• The average delay per job
• Spends waiting in queue plus the service time
• Estimation performed by
• The job arrival rate
• Number of jobs entering the system per unit time
  l
• The job service rate
• Number of jobs the system serves per unit time
  when it is constantly busy L

16
Formalization Littles Theorem
L l W
Average number of jobs in the system jobs
arrival rate x average time a job is spent in the
system
17
Littles Theorem (cont.)

• Holds for many complex arrival-departure systems.
• Crowded systems (large L) are associated with
  long job delays (large W) and reversely.
• Example
• On a rainy day, traffic on a rush hour moves
  slower than average (large W), while the streets
  are more crowded (large L).
• A fast-food restaurant (small W) needs a smaller
  waiting area (small L) than a regular restaurant
  for the same customer arrival rate.

18
Littles Theorem (cont.)

• Example
• If l is the arrival rate in a transmission line,
  Lq is the average number of packets waiting in
  queue (but not under transmission), and Wq is the
  average time spent by a packet waiting in queue
  (not including the transmission time), Littles
  Theorem gives
• Lq l Wq
• Furthermore, if L is the average number of
  packets in the system, and W is the average time
  spent by a packet in the system, then Littles
  Theorem gives the average number of packets in
  the systme as
• L l W

19
Model Queuing System

• Use Queuing models to
• Describe the behavior of queuing systems
• Evaluate system performance

20
Characteristics of queuing systems

• Arrival Process
• The distribution that determines how the tasks
  arrives in the system.
• Service Process
• The distribution that determines the task
  processing time
• Number of Servers
• Total number of servers available to process the
  tasks

21
Kendall Notation 1/2/3(/4/5/6)

• Six parameters in shorthand
• First three typically used, unless specified
• Arrival Distribution
• Service Distribution
• Number of servers
• Total Capacity (infinite if not specified)
• Population Size (infinite)
• Service Discipline (FCFS/FIFO)

22
Distributions

• M stands for "Markovian", implying exponential
  distribution for service times or inter-arrival
  times.
• D Deterministic (e.g. fixed constant)
• Ek Erlang with parameter k
• Hk Hyperexponential with param. k
• G General (anything)

23
Kendall Notation Examples

• M/M/1
• Poisson arrivals and exponential service, 1
  server, infinite capacity and population, FCFS
  (FIFO)
• the simplest realistic queue
• M/M/m
• Same, but M servers
• G/G/3/20/1500/SPF
• General arrival and service distributions, 3
  servers, 17 queue slots (20-3), 1500 total jobs,
  Shortest Packet First

24
Poisson Arrivals MPoisson Process

• For a poisson process with average arrival rate
  , the probability of seeing n arrivals in time
  interval delta t

25
Poisson process Exponential distribution

• Inter-arrival time t (time between arrivals) in a
  Poisson process follows exponential distribution
  with parameter

26
Exponential Service M Exponential distribution

• The server service times have an exponential
  distribution with parameter m. If t is the
  service time of the nth job
• Pr(t) m e - m t
• E(t) 1/ m

27
Exponential distributionImportant character

• The exponential distribution is memoryless
• The additional time needed to complete servicing
  a job in progress is independent of when the
  service started.
• The time up to the next arrival is independent of
  when the previous arrival occurred.

28
Analysis of M/M/1 queue

• Given
• l Arrival rate of jobs (packets on input link)
• m Service rate of the server (output link)
• Solve
• L average number in the system
• Lq average number in the queue
• W average waiting time in whole system
• Wq average waiting time in the queue

29
M/M/1 queue model
30
Solving queuing systems

• 4 unknowns L, Lq W, Wq
• Relationships
• LlW
• LqlWq (steady-state argument)
• W Wq (1/m)
• If we know any one, we can find the others
• Finding L is hard or easy depending on the type
  of system. In general with Pn denoted as the
  steady-state probabilities

31
Analysis of M/M/1 queue

• Goal A closed form expression of the probability
  of the number of jobs in the queue (Pi) given
  only l and m

32
Equilibrium conditions
Define to be the probability of having
n tasks in the system at time t
33
Analysis of M/M/1 queue

• Steady-state probabilities Pn
• The frequency of transitions from n to n1 is
  equal to the frequency of transitions from n1 to
  n.
• The probability that the system is in state n and
  makes a transition to n1 in the next transition
  interval is the same as the probability that the
  system is in state n1 and makes a transition to
  n.

34
Equilibrium conditions
l
l
l
l
n1
n
n-1
m
m
m
m
35
Solving for P0 and Pn

• Step 1
• Step 2

36
Solving for P0 and Pn

• Step 3
• Step 4

For utilization factor ? lt 1, service rate
exceeds arrival rate
37
Solving for L
38
Solving W, Wq and Lq
39
Online M/M/1 animation

• http//www.dcs.ed.ac.uk/home/jeh/Simjava/queueing/
  mm1_q/mm1_q.html

40
Response Time vs. Arrivals
? -gt 1, W -gt 8 for system to be stablize, ? lt 1
41
Stable Region
linear region
42
Example

• On a network gateway, measurements show that the
  packets arrive at a mean rate of 125 packets per
  second (pps) and the gateway takes about 2
  millisecs to forward them. Assuming an M/M/1
  model, what is the probability of buffer overflow
  if the gateway had only 13 buffers. How many
  buffers are needed to keep packet loss below one
  packet per million?

43
Example

• Measurement of a network gateway
• mean arrival rate (l) 125 Packets/s
• mean response time (m) 2 ms
• Assuming exponential arrivals
• What is the gateways utilization?
• What is the probability of n packets in the
  gateway?
• mean number of packets in the gateway?
• The number of buffers so P(overflow) is lt10-6?

44
Example

• Arrival rate ?
• Service rate µ
• Gateway utilization ? ?/µ
• Prob. of n packets in gateway
• Mean number of packets in gateway

45
Example

• Arrival rate ? 125 pps
• Service rate µ 1/0.002 500 pps
• Gateway utilization ? ?/µ 0.25
• Prob. of n packets in gateway
• Mean number of packets in gateway

46
Example

• Probability of buffer overflow
• To limit the probability of loss to less than
  10-6

47
Example

• Probability of buffer overflow P(more than
  13 packets in gateway)
• To limit the probability of loss to less than
  10-6

48
Example

• Probability of buffer overflow P(more than
  13 packets in gateway) P13 P14 P15 P16
  
• ?13 ?14 ?14 ?15 ?15 ?16
• given ? lt 1, P(more than 13 packets in
  gateway)
• ?13 0.2513 1.49x10-8 15 packets per
  billion packets

49
Example

• To limit the probability of loss to less than
  10-6
• or

50
Example

• To limit the probability of loss to less than
  10-6
• or 9.96

51
Empirical Example
M/M/m system