Upcoming SlideShare
×

Queue

2,448
-1

Published on

Published in: Technology, Education
1 Comment
2 Likes
Statistics
Notes
• Full Name
Comment goes here.

Are you sure you want to Yes No
Your message goes here
• thanks you

Are you sure you want to  Yes  No
Your message goes here
Views
Total Views
2,448
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
150
1
Likes
2
Embeds 0
No embeds

No notes for slide

Queue

1. 1. Queuing theory with Applications <ul>Presented By : Rajeev N Bharshetty ( USN : 1RV08IS036 ) </ul>
2. 2. Queuing Theory...What is it? <ul><li>“ The study of the waiting times, lengths, and other properties of queues ” .
3. 3. “ The theoretical study of waiting lines, expressed in mathematical terms--including components such as number of waiting lines, number of servers, average wait time, number of queues or lines, and probabilities of queue times' either increasing or decreasing ” . </li></ul>
4. 4. Why queues form? Queues or waiting lines arise when the demand for a service facility exceeds the capacity of that facility, that is, the customers do not get service immediately upon request but must wait, or the service facilities stand idle and wait for customers. Some customers wait when the total number of customers requiring service exceeds the number of service facilities, some service facilities stand idle when the total number of service facilities exceeds the number of customers requiring service.
5. 5. A Queueing System : An Example <ul><li>The simplest queueing system consists of two components (the queue and the server) and two attributes (the inter-arrival time, i and the service time, t). </li></ul><ul>Server System </ul><ul>Queuing System </ul><ul>Queue </ul><ul>Server </ul><ul>Queuing System </ul>
6. 6. Kendall Notation 1/2/3(/4/5/6) Six parameters in shorthand First three typically used, unless specified <ul><li>Arrival Distribution
7. 7. Service Distribution
8. 8. Number of servers
9. 9. Total Capacity (infinite if not specified)
10. 10. Population Size (infinite)
11. 11. Service Discipline (FCFS/FIFO) </li></ul>
12. 12. <ul>Distributions </ul><ul><li>M: stands for &quot;Markovian&quot;, implying exponential distribution for service times or inter-arrival times.
13. 13. D: Deterministic (e.g. fixed constant)
14. 14. E k : Erlang with parameter k
15. 15. H k : Hyperexponential with param. k
16. 16. G: General (anything) </li></ul>
17. 17. Kendall Notation Examples <ul><li>M/M/1: </li></ul>Poisson arrivals and exponential service, 1 server, infinite capacity and population, FCFS (FIFO) the simplest ‘realistic’ queue <ul><li>M/M/m </li></ul>Same, but M servers <ul><li>G/G/3/20/1500/SPF </li></ul>General arrival and service distributions, 3 servers, 17 queue slots (20-3), 1500 total jobs, Shortest Packet First
18. 18. <ul>Analysis of M/M/1 queue </ul><ul><li>Given: </li><ul><li> : Arrival rate of jobs (packets on input link)
19. 19.  : Service rate of the server (output link) </li></ul><li>Solve: </li><ul><li>L :average number in queuing system
20. 20. L q :average number in the queue
21. 21. W :average waiting time in whole system
22. 22. W q :average waiting time in the queue </li></ul></ul>
23. 23. <ul>M/M/1 queue model </ul><ul> </ul><ul> </ul><ul>W q </ul><ul>W </ul><ul>L </ul><ul>L q </ul>
24. 24. <ul>Solving queuing systems </ul><ul><li>4 unknowns: L, L q W, W q
25. 25. Relationships: </li><ul><li>L=  W
26. 26. L q =  Wq (steady-state argument)
27. 27. W = W q + (1/  ) </li></ul><li>If we know any 1, can find the others
28. 28. Finding L is hard or easy depending on the type of system. In general: </li></ul>
29. 29. <ul>Equilibrium conditions </ul><ul> </ul><ul>n+1 </ul><ul>n </ul><ul>n-1 </ul><ul> </ul><ul> </ul><ul> </ul>
30. 30. <ul>Solving for P 0 and P n </ul><ul><li>Step 1 :
31. 31. Step 2 : </li></ul>
32. 32. <ul>Solving for P 0 and P n </ul><ul><li>Step 3:
33. 33. Step 4 : </li></ul>
34. 34. <ul>Solving for L </ul>
35. 35. <ul>Solving W, W q and L q </ul>
36. 36. <ul>Applications of Queuing Theory </ul><ul><li>Telecommunications
37. 37. Traffic control
38. 38. Determining the sequence of computer operations
39. 39. Predicting computer performance
40. 40. Health services (eg. control of hospital bed assignments)
41. 41. Airport traffic, airline ticket sales
42. 42. Layout of manufacturing systems. </li></ul>
43. 43. Application in Computer Networking <ul><li>Open Queueing Network . </li></ul>
44. 44. Applications <ul><li>Closed Queuing Network </li></ul>
45. 45. Sample Problem <ul><li>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 milliseconds 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? </li></ul>
46. 46. Problem Analysis <ul><li>Measurement of a network gateway: </li><ul><li>mean arrival rate (l): 125 Packets/s
47. 47. mean response time (m): 2 ms </li></ul><li>Assuming exponential arrivals: </li><ul><li>What is the gateway’s utilization?
48. 48. What is the probability of n packets in the gateway?
49. 49. mean number of packets in the gateway?
50. 50. The number of buffers so P(overflow) is <10 -6 ? </li></ul></ul>
51. 51. Solution... <ul><li>Arrival rate λ = 125 pps
52. 52. Service rate μ = 1/0.002 = 500 pps
53. 53. Gateway utilization ρ = λ/μ = 0.25
54. 54. Prob. of n packets in gateway =
55. 55. Mean number of packets in gateway = </li></ul>
56. 56. Solution... Probability of buffer overflow: = P(more than 13 packets in gateway) = ρ 13 = 0.25 13 = 1.49x10 -8 = 15 packets per billion packets To limit the probability of loss to less than 10 -6 :
57. 57. Solution... To limit the probability of loss to less than 10 -6 : or = 9.96
58. 58. <ul><li>Thank You </li></ul>
1. A particular slide catching your eye?

Clipping is a handy way to collect important slides you want to go back to later.