Fundamental Characteristics of Queues with Fluctuating Load (appeared in SIGMETRICS 2006) VARUN GUPTA Joint with: Mor Harchol-Balter Carnegie Mellon Univ.

Slides:



Advertisements
Similar presentations
VARUN GUPTA Carnegie Mellon University 1 With: Mor Harchol-Balter (CMU)
Advertisements

EISCambridge 2009MoN8 Exploring Markov models for gate-limited service and their application to network-based services Glenford Mapp and Dhawal Thakker.
SMA 6304/MIT2.853/MIT2.854 Manufacturing Systems Lecture 19-20: Single-part-type, multiple stage systems Lecturer: Stanley B. Gershwin
Anshul Gandhi (Carnegie Mellon University) Varun Gupta (CMU), Mor Harchol-Balter (CMU) Michael Kozuch (Intel, Pittsburgh)
Page 1 Alan Scheller-Wolf Lunteren, The Netherlands January 15, 2013 Things I Thought I Knew About Queueing Theory, but was Wrong About: Part 1, Multiserver.
Nur Aini Masruroh Queuing Theory. Outlines IntroductionBirth-death processSingle server modelMulti server model.
#11 QUEUEING THEORY Systems Fall 2000 Instructor: Peter M. Hahn
ECS 152A Acknowledgement: slides from S. Kalyanaraman & B.Sikdar
A gentle introduction to fluid and diffusion limits for queues Presented by: Varun Gupta April 12, 2006.
Mean Delay in M/G/1 Queues with Head-of-Line Priority Service and Embedded Markov Chains Wade Trappe.
20. Extinction Probability for Queues and Martingales
FINDING THE OPTIMAL QUANTUM SIZE Revisiting the M/G/1 Round-Robin Queue VARUN GUPTA Carnegie Mellon University.
25 June 2015Comp 122, Spring 2004 Asymptotic Notation, Review of Functions & Summations.
1 Alan Scheller-Wolf Joint with: Mor Harchol-Balter, Taka Osogami, Adam Wierman, and Li Zhang. Dimensionality Reduction for the analysis of Cycle Stealing,
Join-the-Shortest-Queue (JSQ) Routing in Web Server Farms
Fundamental Characteristics of Queues with Fluctuating Load VARUN GUPTA Joint with: Mor Harchol-Balter Carnegie Mellon Univ. Alan Scheller-Wolf Carnegie.
Fluid level in tandem queues with an On/Off source VARUN GUPTA Carnegie Mellon University Joint work with PETER HARRISON Imperial College.
1 TCOM 501: Networking Theory & Fundamentals Lectures 9 & 10 M/G/1 Queue Prof. Yannis A. Korilis.
Effect of higher moments of job size distribution on the performance of an M/G/k system VARUN GUPTA Joint work with: Mor Harchol-Balter Carnegie Mellon.
Effect of higher moments of job size distribution on the performance of an M/G/k system VARUN GUPTA Joint work with: Mor Harchol-Balter Carnegie Mellon.
Lecture 14 – Queuing Systems
1 Real-Time Queueing Network Theory Presented by Akramul Azim Department of Electrical and Computer Engineering University of Waterloo, Canada John P.
1 Chapters 9 Self-SimilarTraffic. Chapter 9 – Self-Similar Traffic 2 Introduction- Motivation Validity of the queuing models we have studied depends on.
RAQFM – a Resource Allocation Queueing Fairness Measure David Raz School of Computer Science, Tel Aviv University Jointly with Hanoch Levy, Tel Aviv University.
Module 2: Representing Process and Disturbance Dynamics Using Discrete Time Transfer Functions.
Copyright warning. COMP5348 Lecture 6: Predicting Performance Adapted with permission from presentations by Alan Fekete.
Decentralised load balancing in closed and open systems A. J. Ganesh University of Bristol Joint work with S. Lilienthal, D. Manjunath, A. Proutiere and.
1 Exponential distribution: main limitation So far, we have considered Birth and death process and its direct application To single server queues With.
Imaginary & Complex Numbers 5-3 English Casbarro Unit 5: Polynomials.
2.1 Computational Tractability. 2 Computational Tractability Charles Babbage (1864) As soon as an Analytic Engine exists, it will necessarily guide the.
Polynomial Functions and Their Graphs
Lecture 10: Queueing Theory. Queueing Analysis Jobs serviced by the system resources Jobs wait in a queue to use a busy server queueserver.
NETE4631:Capacity Planning (2)- Lecture 10 Suronapee Phoomvuthisarn, Ph.D. /
Analytical modeling of part supply process in a bin-kanban system with logistic trains Fabio Bursi, Elisa Gebennini, Andrea Grassi, Bianca Rimini.
Introduction to Queueing Theory
Solving Quadratic Equations by Factoring. Solution by factoring Example 1 Find the roots of each quadratic by factoring. factoring a) x² − 3x + 2 b) x².
Squared coefficient of variation
1 Elements of Queuing Theory The queuing model –Core components; –Notation; –Parameters and performance measures –Characteristics; Markov Process –Discrete-time.
Modeling and Analysis of Computer Networks
OMG Operations Management Spring 1997 CLASS 4: THE IMPACT OF VARIABILITY Harry Groenevelt.
Appointment Systems - a Stochastic and Fluid Approach Michal Penn The William Davidson Faculty of Industrial Engineering and Management Technion - Israel.
An Optimal Design of the M/M/C/K Queue for Call Centers
State N 2.6 The M/M/1/N Queueing System: The Finite Buffer Case.
The M/M/ N / N Queue etc COMP5416 Advanced Network Technologies.
OPERATING SYSTEMS CS 3530 Summer 2014 Systems and Models Chapter 03.
Asymptotics and Recurrence Equations Prepared by John Reif, Ph.D. Analysis of Algorithms.
Lec 6. Second Order Systems
PreCalculus Section P.1 Solving Equations. Equations and Solutions of Equations An equation that is true for every real number in the domain of the variable.
Queuing Theory.  Queuing Theory deals with systems of the following type:  Typically we are interested in how much queuing occurs or in the delays at.
1 Task Assignment with Unknown Duration Mor Harchol-Balter Carnegie Mellon.
Chapter 5 Elementary Stochastic Analysis Prof. Ali Movaghar.
Flows and Networks Plan for today (lecture 3): Last time / Questions? Output simple queue Tandem network Jackson network: definition Jackson network: equilibrium.
Random Variables r Random variables define a real valued function over a sample space. r The value of a random variable is determined by the outcome of.
QUEUING. CONTINUOUS TIME MARKOV CHAINS {X(t), t >= 0} is a continuous time process with > sojourn times S 0, S 1, S 2,... > embedded process X n = X(S.
Managerial Decision Making Chapter 13 Queuing Models.
System Provisioning.
Equations Quadratic in form factorable equations
Load Balancing and Data centers
Flows and Networks Plan for today (lecture 4):
Queuing Theory Non-Markov Systems
Time-dependent queue modelling
Lecture on Markov Chain
Trigonometric Identities
Using Factoring To Solve
Equations Quadratic in form factorable equations
Erlang, Hyper-exponential, and Coxian distributions
Kendall’s Notation ❚ Simple way of summarizing the characteristics of a queue. Arrival characteristics / Departure characteristics / Number of servers.
CS723 - Probability and Stochastic Processes
Presentation transcript:

Fundamental Characteristics of Queues with Fluctuating Load (appeared in SIGMETRICS 2006) VARUN GUPTA Joint with: Mor Harchol-Balter Carnegie Mellon Univ. Alan Scheller-Wolf Carnegie Mellon Univ. Uri Yechiali Tel Aviv Univ.

2 Motivation Clients Server Farm Requests

3 Motivation Clients Server Farm Requests

4 Motivation Clients Server Farm Requests

5 Motivation Clients Server Farm Requests

6 Motivation Clients Server Farm Requests

7 Motivation Clients Server Farm Requests

8 Motivation Clients Server Farm Requests

9 Motivation Clients Server Farm Requests Real World  Fluctuating arrival and service intensities

10 A Simple Model HH LL exp(  H ) exp(  L ) High Load Low Load

11 Poisson Arrivals Exponential Job Size Distribution H /  H > L /  L H >  H possible, only need stability A Simple Model High Load Low Load  H,  H L,  L exp(  ) H HH L LL

12 The Markov Chain Phase Number of jobs L H H HH  LL L  H HH LL L...  Solving the Markov chain provides no behavioral insight

13  H HH L LL N = Number of jobs in the fluctuating load system Lets try approximating N using (simpler) non- fluctuating systems

14  H HH L LL Method 1 N mix

15 H HH L LL Q: Is N mix ≈ N? A: Only when   0 Method 1 N mix ½½ ½½ +   ,

16  H HH L LL Method 2

17 avg( H, L ) avg(  H,  L ) Method 2 ≡ N avg Q: Is N avg ≈ N? A: When      ,

18 Example    H =1, H =0.99  L =1, L =0.01 E[N mix ] ≈ 49.5 E[N avg ] = 1 00 

19 Observations Fluctuating system can be worse than non- fluctuating   0 and    asymptotes can be very far apart E[N mix ] > E[N avg ] E[N mix ]  E[N avg ]

20 Questions Is fluctuation always bad? Is E[N] monotonic in  ? Is there a simple closed form approximation for E[N] for intermediate  ’s? How do queue lengths during High Load and Low Load phase compare? How do they compare with N avg ? More than 40 years of research has not addressed such fundamental questions!

21 Outline  Is E[N mix ] ≥ E[N avg ], always?  Is E[N] monotonic in  ?  Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

22 Prior Work Fluid/Diffusion Approximations Transforms Matrix Analytic & Spectral Analysis - P. Harrison - Adan and Kulkarni Numerical Approaches Involves solution of cubic - Clarke - Neuts - Yechiali and Naor Involves solution of cubic - Massey - Newell - Abate, Choudhary, Whitt Limiting Behavior But cubic equations have a close form solution… ?

23 Good luck understanding this!

24 Asymptotics for E[N] ( H <  H ) E[N avg ] E[N mix ] E[N]  (switching rate) High fluctuation    H =1, H =0.99  L =1, L =0.01 E[N mix ] > E[N avg ]  Low fluctuation

25 Asymptotics for E[N] ( H <  H )  E[N] E[N mix ] E[N avg ] Agrees with our example (  H =  L ) Ross’s conjecture for systems with constant service rate: “Fluctuation increases mean delay” Q: Is this behavior possible? A: Yes  E[N] E[N avg ] E[N mix ]

26 Our Results  E[N] (  H - H ) > (  L - L ) (  H - H ) = (  L - L ) (  H - H ) < (  L - L ) Define the slacks during L and H as s L =  L - L s H =  H - H  E[N] 

27 Our Results Define the slacks during L and H as s L =  L - L s H =  H - H Not load but slacks determine the response times! s H > s L s H = s L s H < s L KEY IDEA  E[N]  

28 Outline  Is E[N mix ] ≥ E[N avg ], always?  Is E[N] monotonic in  ?  Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

29 Outline Is E[N mix ] ≥ E[N avg ], always? No  Is E[N] monotonic in  ?  Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

30 Monotonicity of E[N]  : Mean Queue Length H L  ’ : Mean Queue Length H L

31 Monotonicity of E[N] We show : E[N] is monotonic in   : Mean Queue Length H L  ’ : Mean Queue Length H L Not obvious that true for all ,  ’ with  ’<  !

32 Outline Is E[N mix ] ≥ E[N avg ], always? No  Is E[N] monotonic in  ?  Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

33 Outline Is E[N mix ] ≥ E[N avg ], always? No Is E[N] monotonic in  ? Yes  Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

34 Approximating E[N] Express the first moment as * E[N] = E[N mix ]r+E[N avg ](1-r) Approximate r by the root of a quadratic KEY IDEA * True for H <  H ; a similar expression exists for case of transient overload

35 Approximating E[N]  E[N] Exact Approx.  H =  L =1, H =0.95, L =0.2

36 Approximating E[N]  E[N] Exact Approx.  H =  L =1, H =0.95, L =0.2

37 Approximating E[N]  Exact Approx.  H =  L =1, H =1.2, L = E[N]

38 Approximating E[N]  Exact Approx.  H =  L =1, H =1.2, L = E[N]

39 Outline Is E[N mix ] ≥ E[N avg ], always? No Is E[N] monotonic in  ? Yes  Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

40 Outline Is E[N mix ] ≥ E[N avg ], always? No Is E[N] monotonic in  ? Yes Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

41 Stochastic Ordering refresher For random variables X and Y X  st Y  Pr{X  i}  Pr{Y  i} for all i. X  st Y  E[f(X)]  E[f(Y)] for all increasing f –E[X k ]  E[Y k ] for all k  0.

42 Notation N H : Number of jobs in system during H phase N L : Number of jobs in system during L phase N = (N H +N L )/2 H,  H L,  L exp(  ) NHNH NLNL

43 Stochastic Orderings for N L, N H N L ≥ st N M/M/1/L N H ≤ st N M/M/1/H N H ≥ st N L N H ≥ st N avg N L  st N avg ? ? ? ? ? H,  H L,  L exp(  ) NHNH NLNL N H increases stochastically as  ↓ Conjecture:

44 Outline Is E[N mix ] ≥ E[N avg ], always? No Is E[N] monotonic in  ? Yes Simple closed form approximation for E[N]  Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

45 Outline Is E[N mix ] ≥ E[N avg ], always? No Is E[N] monotonic in  ? Yes Simple closed form approximation for E[N] Stochastic orderings for the number of jobs seen by an arrival into an H phase, L phase  Application: Capacity Planning

46 Scenario Application: Capacity Provisioning  H HH L LL  2 H HH 2 L LL Aim: To keep the mean response times same

47 Scenario Application: Capacity Provisioning  H HH L LL  2 H 2H2H 2 L 2L2L Question: What is the effect of doubling the arrival and service rates on the mean response time?

48 What happens to the mean response time when,  are doubled in the fluctuating load queue? Halves Remains almost the sameReduces by less than half Reduces by more than half  A:  D:  C:  B:

49 What happens to the mean response time when,  are doubled in the fluctuating load queue? Halves Remains almost the sameReduces by less than half Reduces by more than half  A:  D:  C:  B:

50 What happens to the mean response time when,  are doubled in the fluctuating load queue? Halves Remains almost the sameReduces by less than half Reduces by more than half  A:  D:  C:  B: Look at slacks! A: s H = s L B: s H > s L C: s H < s L D: s H < 0,   0  reduces by half  more than half  less than half  remains same

51 Our Contributions Give a simple characterization of the behavior of E[N] vs.  Provide simple (and tight) quadratic approximations for E[N] Prove the first stochastic ordering results for the fluctuating load model

52 Thank you

53 Analysis of E[N] First steps: –Note that it suffices to look at switching points –Express N L = f(N H ) N H = g(N L ) –The problem reduces to finding  Pr{N H =0} and Pr{N L =0} H,  H L,  L NHNH NLNL N L =f(g(N L )) f g

54 –Find the root  of a cubic (the characteristic matrix polynomial in the Spectral Expansion method) –Express E[N] in terms of  E[N] = The simple way forward… H,  H L,  L f g A  A - A  H (  L - L )  0 H +  L (  H - H )  0 L - (  L - L )(  H - H ) 2  (  A - A ) + Where  0 L =  0 H =  (  A - A )   L (  -1)(  H  - H )  (  A - A )   H (  -1)(  L  - L ) NHNH NLNL Difficult to even prove the monotonicity of E[N] wrt  using this!

55 Our approach (contd.) Express the first moment as E[N] = f 1 (  )r+f 0 (  )(1-r) –r is the root of a (different) cubic –r  1 as  0 and r  0 as  KEY IDEA

56 Monotonicity of E[N] E[N] = f 1 (  )r+f 0 (  )(1-r) r is monotonic in   E[N] is monotonic in  The cubic for r has maximum power of  as r 

57 Monotonicity of E[N] E[N] = f 1 (  )r+f 0 (  )(1-r) r is monotonic in   E[N] is monotonic in  The cubic for r has maximum power of  as r  Need at least 3 roots for  when r=c 1 but  has at most 2 roots c1c1

58 Monotonicity of E[N] E[N] = f 1 (  )r+f 0 (  )(1-r) r is monotonic in   E[N] is monotonic in  The cubic for r has maximum power of  as r  Need at least 2 positive roots for  when r=c 2 but for r>1 product of roots is negative c2c2

59 Monotonicity of E[N] E[N] = f 1 (  )r+f 0 (  )(1-r) r is monotonic in   E[N] is monotonic in  The cubic for r has maximum power of  as r  E[N] is monotonic in  !

60 Why do slacks matter? Fact: The mean response time in an M/M/1 queue is (  - ) -1 –Higher slacks  Lower mean response times What is the fraction of customers departing during H H,  H L,  L exp(  ) when  ? H,  H L,  L exp(  ) when  0? HH  H +  L H H + L ?

61 Why do slacks matter? when  ? H,  H L,  L exp(  ) when  0? HH  H +  L H H + L ? Fact: The mean response time in an M/M/1 queue is (  - ) -1 –Higher slacks  Lower mean response times What is the fraction of customers departing during H H,  H L,  L exp(  )

62 Why do slacks matter? when  ? H,  H L,  L exp(  ) when  0? AA HH ?  Fact: The mean response time in an M/M/1 queue is (  - ) -1 –Higher slacks  Lower mean response times What is the fraction of customers departing during H As switching rates decrease, larger fraction of customers experience lower mean response times when s H >s L H,  H L,  L exp(  )

63 Q: What happens to E[N] when we double ’s and  ’s? A: System A:, ,  System B: 2, 2 ,  ?

64 Q: What happens to E[N] when we double ’s and  ’s? A: System A:, ,  System B: 2, 2 ,  System C: 2, 2 , 2  E[N] remains same in going from A to C A) s L = s H : remains same B) s L > s H : increases, but by less than twice C) s L < s H : decreases D)  0,  H >1 : queue lengths become twice as switching rates halve, E[N] doubles