1 Multiple class queueing networks Mean Value Analysis - Open queueing networks - Closed queueing networks.

Slides:



Advertisements
Similar presentations
Closed Queuing Networks (Mean Value Analysis). Closed Queuing Networks Arise in two situations Arise in two situations When “source” of requests is explicitly.
Advertisements

Performance Evaluation: Markov Models, revisited CSCI 8710 E. Kraemer.
10/12/1999(c) Ian Davis1 Predicting performance Topics: –Operational analysis of network models –Markov models.
Workload Characterization Sept. 23 rd, 2008 CSCI 8710.
Ó 1998 Menascé & Almeida. All Rights Reserved.1 Part IV Capacity Planning Methodology.
Performance Engineering Methodology Chapter 4. Performance Engineering Performance engineering analyzes the expected performance characteristics of a.
1 Part IV Capacity Planning Methodology © 1998 Menascé & Almeida. All Rights Reserved.
Queuing Analysis Based on noted from Appendix A of Stallings Operating System text 6/10/20151.
1 Part II Web Performance Modeling: basic concepts © 1998 Menascé & Almeida. All Rights Reserved.
System Performance & Scalability i206 Fall 2010 John Chuang.
Queueing Systems They’re EVERYWHERE!. Basic Concept Service Request... Done Numerous requests made for service Lines back up waiting for service Waiting.
CSE 221: Probabilistic Analysis of Computer Systems Topics covered: Simple queuing models (Sec )
Single queue modeling. Basic definitions for performance predictions The performance of a system that gives services could be seen from two different.
Computer Networks: Performance Measures1 Computer Network Performance Measures.
Data Communication and Networks Lecture 13 Performance December 9, 2004 Joseph Conron Computer Science Department New York University
Multiple Tiers in Action
Queueing Network Model. Single Class Model Open - Infinite stream of arriving customers Closed - Finite population eg Intranet users Indistinguishable.
Measuring Performance Chapter 12 CSE807. Performance Measurement To assist in guaranteeing Service Level Agreements For capacity planning For troubleshooting.
Networks: Performance Measures1 Network Performance Measures.
1 Internet Management and Security We will look at management and security of networks and systems. Systems: The end nodes of the Internet Network: The.
On G-network and resource allocation in multimedia systems 報告者 : 王敬育.
7/3/2015© 2007 Raymond P. Jefferis III1 Queuing Systems.
Operational Laws A large number of day-to-day problem in computer systems performance analysis can be solved by using some simple relationships that do.
1 Part VI System-level Performance Models for the Web © 1998 Menascé & Almeida. All Rights Reserved.
Using Standard Industry Benchmarks Chapter 7 CSE807.
Queuing Networks. Input source Queue Service mechanism arriving customers exiting customers Structure of Single Queuing Systems Note: 1.Customers need.
Computer System Lifecycle Chapter 1. Introduction Computer System users, administrators, and designers are all interested in performance evaluation. Whether.
Computer Networks Performance Evaluation. Chapter 12 Single Class MVA Performance by Design: Computer Capacity Planning by Example Daniel A. Menascé,
yahoo.com SUT-System Level Performance Models yahoo.com SUT-System Level Performance Models8-1 chapter12 Single Class MVA.
From Systems to Descriptive Models Chapter 2. What need to know Modeling Graphical Notation for Queues Example Open workload Multiclass DB example Open.
Performance of Web Applications Introduction One of the success-critical quality characteristics of Web applications is system performance. What.
AN INTRODUCTION TO THE OPERATIONAL ANALYSIS OF QUEUING NETWORK MODELS Peter J. Denning, Jeffrey P. Buzen, The Operational Analysis of Queueing Network.
M EAN -V ALUE A NALYSIS Manijeh Keshtgary O VERVIEW Analysis of Open Queueing Networks Mean-Value Analysis 2.
Database Architecture Introduction to Databases. The Nature of Data Un-structured Semi-structured Structured.
Queuing models Basic definitions, assumptions, and identities Operational laws Little’s law Queuing networks and Jackson’s theorem The importance of think.
Modeling and Performance Evaluation of Network and Computer Systems Introduction (Chapters 1 and 2) 10/4/2015H.Malekinezhad1.
E-Commerce Architectures and Technologies Rob Oshana Southern Methodist University.
Computer Networks Performance Metrics. Performance Metrics Outline Generic Performance Metrics Network performance Measures Components of Hop and End-to-End.
NETE4631:Capacity Planning (2)- Lecture 10 Suronapee Phoomvuthisarn, Ph.D. /
1 CS 501 Spring 2006 CS 501: Software Engineering Lecture 22 Performance of Computer Systems.
CSE 538 Instructor: Roch Guerin – Office hours: Mon & Wed 4:00-5:00pm in Bryan 509 TA: Junjie Liu
Analytic Evaluation of Quality of Service for On-Demand Data Delivery Hongfei Guo Haonan Tan
1 Chapters 8 Overview of Queuing Analysis. Chapter 8 Overview of Queuing Analysis 2 Projected vs. Actual Response Time.
Analytic Evaluation of Shared-Memory Systems with ILP Processors Daniel J. Sorin, Vijay S. Pai, Sarita V. Adve, Mary K. Vernon, and David A. Wood Presented.
1 Challenges in Scaling E-Business Sites  Menascé and Almeida. All Rights Reserved. Daniel A. Menascé Department of Computer Science George Mason.
Network Design and Analysis-----Wang Wenjie Queuing Theory III: 1 © Graduate University, Chinese academy of Sciences. Network Design and Performance Analysis.
Queueing Models with Multiple Classes CSCI 8710 Tuesday, November 28th Kraemer.
Review. Why can average request time be used to determine the work-load required for saturation?
1 Components performance modelling - Outline of queue networks - Mean Value Analisys (MVA) for open and close queue networks.
1 Part VII Component-level Performance Models for the Web © 1998 Menascé & Almeida. All Rights Reserved.
NETE4631: Network Information System Capacity Planning (2) Suronapee Phoomvuthisarn, Ph.D. /
Capacity Planning Plans Capacity Planning Operational Laws
Ó 1998 Menascé & Almeida. All Rights Reserved.1 Part VI System-level Performance Models for the Web (Book, Chapter 8)
Review Session Jehan-François Pâris. Agenda Statistical Analysis of Outputs Operational Analysis Case Studies Linear Regression.
Internet Applications: Performance Metrics and performance-related concepts E0397 – Lecture 2 10/8/2010.
Ó 1998 Menascé & Almeida. All Rights Reserved.1 Part II System Performance Modeling: basic concepts, operational analysis (book, chap. 3)
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.
QUEUING THEORY 1.  - means the number of arrivals per second   - service rate of a device  T - mean service time for each arrival   = ( ) Utilization,
Queuing Theory Simulation & Modeling.
Ó 1998 Menascé & Almeida. All Rights Reserved.1 Part VIII Web Performance Modeling (Book, Chapter 10)
Ó 1998 Menascé & Almeida. All Rights Reserved.1 Part VI System-level Performance Models for the Web.
Software Architecture in Practice Theoretical Models for Performance.
Queuing Theory Queuing Theory.
B.Ramamurthy Appendix A
Computer Network Performance Measures
Queuing models Basic definitions, assumptions, and identities
Queuing models Basic definitions, assumptions, and identities
Computer Network Performance Measures
Queuing Theory By: Brian Murphy.
Queuing Analysis.
Presentation transcript:

1 Multiple class queueing networks Mean Value Analysis - Open queueing networks - Closed queueing networks

2 Open queueing network DISK CD outgoing requests incoming requests CPU DISK CD CPU M clients Closed queueing network (number finite of users)

3 Incoming request class Different kind of requests should be in a system (queueing network) that need different services by the servers, i.e: -a database server is subject to two type of transactions: -simple query (that needs only read activities on the disks) -updating transactions (that needs read and write activities on the disks) -a web server is subject to two type of requests: -Read of a little file -Uploading of a big file

4 Definitions K: number of queues i: queue identification r: class identification (from 1 to R)  r : arrival rate for class r request = ( 1, 2,..., R ) V i,r : average number of visits a class r request makes to server i from its generation to its service time (request goes out from the system if open network)

Definitions S i,r : average class r request service time at the server i W i,r : average class r request waiting time in the queue i R i,r : average class r request answer time in the queue i R i,r = S i,r + W i,r

6 Definitions R ’ i,r : average class r request residence time in the queue i from its creation to its service completion time (request goes out from the system in case of open network) R ’ i,r = V i,r R i,r D i,r : request class r service demand to a server in a queue i from its creation to its service completion time (request goes out from the system in case of open network) D i,r = V i,r S i,r

Input parameters D i,r, r Equations. U i,r ( ) = r V i,r S i,r = r D i,r. U i ( ) =  R r=1 U i,r ( ) total utilization factor. R ’ i,r ( ) = D i,r delaying resource R ’ i,r ( ) = D i,r / (1-U i ( )) queuing resource Formulas for multiple class open QNs

. R 0,r ( ) =  K i=1 R ’ i,r ( ). n i,r ( ) = U i,r ( ) / (1-U i ( )) NOTE: total utilization in the denominator. n i, ( ) =  R r=1 n i,r ( ) Formulas for multiple class open QNs

9 DB Server (example ) Class 1 trx: query  = 5 requests per second (tps) D CPU = 0,1 sec Service demand at CPU D DISK1 = 0.08 Service demand at disk 1 D DISK2 = 0.07 Service demand at disk 2 Class 1 trx: updating trx  = 2 requests per second (tps) D CPU = 0,15 sec Service demand at CPU D DISK1 = 0.20 Service demand at disk 1 D DISK2 = 0.10 Service demand at disk 2 CPU DISK1 DISK2

10 DB Server (example ) CPU DISK1 DISK2 Service demand x QueryUpdates CPU0,10,15 DISK10,080,20 DISK20,070,10

11 Utilizations (%) CPU5030 Disk140 Disk Residence times (sec) CPU0, Disk10,401,00 Disk 20,0160,22 Response times (sec)1,061,97

12 DISK TAPE CPU M clients Multiclass closed queue networks (finite number of users)

Notations N r : fixed number of requests in the system for each class (r) N: (N 1, N 2,..., N R ) 1 r : vector where all components are zero except for the r-th component, which is equal to 1

Formulas -> Residence Time Equation for class r R ’ i,r (N)= D i,r [1+n i (N – 1 r )] -> Throughput equation for class r X 0,r = N r /  K r=1 R ’ i,r (N) -> Queue lenght equation for class r n i,r (N) = X 0,r (N) R ’ i,r -> Queue equation n i (N)=  R r=1 n i,r (N)

Example with 2 classes Residence Time Equation for class r R ’ i,r (N)= D i,r [1+n i (N – 1 r )] for example, to evaluate the formulas, when the state is N=(3,4), i.e. 3 customers of class 1 and 4 customers of class 2, we need to know: the average number of users in queue i when there are 2 customers of class 1 and 4 customer of class 2 the average number of users in queue i when there are 3 customers of class 1 and 4 customer of class 3 R ’ i,1 (3,4)= D i,r [1+n i (2,4)] R ’ i,2 (3,4)= D i,r [1+n i (3,3)]