Enviar búsqueda
Cargar
Load models ppt
•
0 recomendaciones
•
783 vistas
Manoj Pathak
Seguir
Educación
Denunciar
Compartir
Denunciar
Compartir
1 de 16
Descargar ahora
Descargar para leer sin conexión
Recomendados
Bab 2 Rekayasa Perangkat Lunak 3
Bab 2 Rekayasa Perangkat Lunak 3
Dimara Hakim
Review on cost estimation technque for web application [part 1]
Review on cost estimation technque for web application [part 1]
Sayed Mohsin Reza
Protoytyping Model
Protoytyping Model
university of education,Lahore
A new model for software costestimation
A new model for software costestimation
ijfcstjournal
Approachesppt 111112121701 Phpapp02
Approachesppt 111112121701 Phpapp02
Archana Survase
Freenome's Biological Machine Learning Platform
Freenome's Biological Machine Learning Platform
Brandon White
Notes on Deploying Machine-learning Models at Scale
Notes on Deploying Machine-learning Models at Scale
Deep Kayal
Naveen_Resume1
Naveen_Resume1
Naveen K K
Recomendados
Bab 2 Rekayasa Perangkat Lunak 3
Bab 2 Rekayasa Perangkat Lunak 3
Dimara Hakim
Review on cost estimation technque for web application [part 1]
Review on cost estimation technque for web application [part 1]
Sayed Mohsin Reza
Protoytyping Model
Protoytyping Model
university of education,Lahore
A new model for software costestimation
A new model for software costestimation
ijfcstjournal
Approachesppt 111112121701 Phpapp02
Approachesppt 111112121701 Phpapp02
Archana Survase
Freenome's Biological Machine Learning Platform
Freenome's Biological Machine Learning Platform
Brandon White
Notes on Deploying Machine-learning Models at Scale
Notes on Deploying Machine-learning Models at Scale
Deep Kayal
Naveen_Resume1
Naveen_Resume1
Naveen K K
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
Splunk
Productionizing Predictive Analytics using the Rendezvous Architecture - for ...
Productionizing Predictive Analytics using the Rendezvous Architecture - for ...
danielschulz2005
Enterprise performance engineering solutions
Enterprise performance engineering solutions
Infosys
Data Architecture Success Stories
Data Architecture Success Stories
Embarcadero Technologies
CV_Mike Yan
CV_Mike Yan
Shiping Yan
Il paradigma DevOps e Continuous Delivery Automation
Il paradigma DevOps e Continuous Delivery Automation
HP Enterprise Italia
Chapter 2_Process Models sunorgamisedASE_finalised.ppt
Chapter 2_Process Models sunorgamisedASE_finalised.ppt
Bule Hora University
Primer on application_performance_modelling_v0.1
Primer on application_performance_modelling_v0.1
Trevor Warren
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
Splunk
AnalyticOps: Lessons Learned Moving Machine-Learning Algorithms to Production...
AnalyticOps: Lessons Learned Moving Machine-Learning Algorithms to Production...
Robert Grossman
Cloud Migration - The Earlier You Instrument, The Faster You Go
Cloud Migration - The Earlier You Instrument, The Faster You Go
Kevin Downs
Unit Testing Essay
Unit Testing Essay
Dani Cox
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
Splunk
B Fn As
B Fn As
Sangwon Ko
WE-06-Testing.ppt
WE-06-Testing.ppt
javed281701
Ijetcas14 413
Ijetcas14 413
Iasir Journals
JAMAL_RESUME
JAMAL_RESUME
Jamal Ouazzani
JAMAL_RESUME
JAMAL_RESUME
Jamal Ouazzani
Ml ops intro session
Ml ops intro session
Avinash Patil
SOLIDWORKS reseller Whitepaper by Promedia Systems
SOLIDWORKS reseller Whitepaper by Promedia Systems
Cavien Clever
Grant Readiness 101 TechSoup and Remy Consulting
Grant Readiness 101 TechSoup and Remy Consulting
TechSoup
mini mental status format.docx
mini mental status format.docx
PoojaSen20
Más contenido relacionado
Similar a Load models ppt
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
Splunk
Productionizing Predictive Analytics using the Rendezvous Architecture - for ...
Productionizing Predictive Analytics using the Rendezvous Architecture - for ...
danielschulz2005
Enterprise performance engineering solutions
Enterprise performance engineering solutions
Infosys
Data Architecture Success Stories
Data Architecture Success Stories
Embarcadero Technologies
CV_Mike Yan
CV_Mike Yan
Shiping Yan
Il paradigma DevOps e Continuous Delivery Automation
Il paradigma DevOps e Continuous Delivery Automation
HP Enterprise Italia
Chapter 2_Process Models sunorgamisedASE_finalised.ppt
Chapter 2_Process Models sunorgamisedASE_finalised.ppt
Bule Hora University
Primer on application_performance_modelling_v0.1
Primer on application_performance_modelling_v0.1
Trevor Warren
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
Splunk
AnalyticOps: Lessons Learned Moving Machine-Learning Algorithms to Production...
AnalyticOps: Lessons Learned Moving Machine-Learning Algorithms to Production...
Robert Grossman
Cloud Migration - The Earlier You Instrument, The Faster You Go
Cloud Migration - The Earlier You Instrument, The Faster You Go
Kevin Downs
Unit Testing Essay
Unit Testing Essay
Dani Cox
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
Splunk
B Fn As
B Fn As
Sangwon Ko
WE-06-Testing.ppt
WE-06-Testing.ppt
javed281701
Ijetcas14 413
Ijetcas14 413
Iasir Journals
JAMAL_RESUME
JAMAL_RESUME
Jamal Ouazzani
JAMAL_RESUME
JAMAL_RESUME
Jamal Ouazzani
Ml ops intro session
Ml ops intro session
Avinash Patil
SOLIDWORKS reseller Whitepaper by Promedia Systems
SOLIDWORKS reseller Whitepaper by Promedia Systems
Cavien Clever
Similar a Load models ppt
(20)
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
Productionizing Predictive Analytics using the Rendezvous Architecture - for ...
Productionizing Predictive Analytics using the Rendezvous Architecture - for ...
Enterprise performance engineering solutions
Enterprise performance engineering solutions
Data Architecture Success Stories
Data Architecture Success Stories
CV_Mike Yan
CV_Mike Yan
Il paradigma DevOps e Continuous Delivery Automation
Il paradigma DevOps e Continuous Delivery Automation
Chapter 2_Process Models sunorgamisedASE_finalised.ppt
Chapter 2_Process Models sunorgamisedASE_finalised.ppt
Primer on application_performance_modelling_v0.1
Primer on application_performance_modelling_v0.1
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
AnalyticOps: Lessons Learned Moving Machine-Learning Algorithms to Production...
AnalyticOps: Lessons Learned Moving Machine-Learning Algorithms to Production...
Cloud Migration - The Earlier You Instrument, The Faster You Go
Cloud Migration - The Earlier You Instrument, The Faster You Go
Unit Testing Essay
Unit Testing Essay
Machine Learning and Analytics Breakout Session
Machine Learning and Analytics Breakout Session
B Fn As
B Fn As
WE-06-Testing.ppt
WE-06-Testing.ppt
Ijetcas14 413
Ijetcas14 413
JAMAL_RESUME
JAMAL_RESUME
JAMAL_RESUME
JAMAL_RESUME
Ml ops intro session
Ml ops intro session
SOLIDWORKS reseller Whitepaper by Promedia Systems
SOLIDWORKS reseller Whitepaper by Promedia Systems
Último
Grant Readiness 101 TechSoup and Remy Consulting
Grant Readiness 101 TechSoup and Remy Consulting
TechSoup
mini mental status format.docx
mini mental status format.docx
PoojaSen20
Employee wellbeing at the workplace.pptx
Employee wellbeing at the workplace.pptx
NirmalaLoungPoorunde1
Measures of Central Tendency: Mean, Median and Mode
Measures of Central Tendency: Mean, Median and Mode
Thiyagu K
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
Marc Dusseiller Dusjagr
A Critique of the Proposed National Education Policy Reform
A Critique of the Proposed National Education Policy Reform
Chameera Dedduwage
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
ssuser54595a
Solving Puzzles Benefits Everyone (English).pptx
Solving Puzzles Benefits Everyone (English).pptx
OH TEIK BIN
Introduction to ArtificiaI Intelligence in Higher Education
Introduction to ArtificiaI Intelligence in Higher Education
pboyjonauth
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
Krashi Coaching
How to Make a Pirate ship Primary Education.pptx
How to Make a Pirate ship Primary Education.pptx
manuelaromero2013
Organic Name Reactions for the students and aspirants of Chemistry12th.pptx
Organic Name Reactions for the students and aspirants of Chemistry12th.pptx
VS Mahajan Coaching Centre
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
9953056974 Low Rate Call Girls In Saket, Delhi NCR
_Math 4-Q4 Week 5.pptx Steps in Collecting Data
_Math 4-Q4 Week 5.pptx Steps in Collecting Data
JhengPantaleon
Introduction to AI in Higher Education_draft.pptx
Introduction to AI in Higher Education_draft.pptx
pboyjonauth
Model Call Girl in Bikash Puri Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Bikash Puri Delhi reach out to us at 🔝9953056974🔝
9953056974 Low Rate Call Girls In Saket, Delhi NCR
Micromeritics - Fundamental and Derived Properties of Powders
Micromeritics - Fundamental and Derived Properties of Powders
ChitralekhaTherkar
Hybridoma Technology ( Production , Purification , and Application )
Hybridoma Technology ( Production , Purification , and Application )
Sakshi Ghasle
Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)
eniolaolutunde
Separation of Lanthanides/ Lanthanides and Actinides
Separation of Lanthanides/ Lanthanides and Actinides
FatimaKhan178732
Último
(20)
Grant Readiness 101 TechSoup and Remy Consulting
Grant Readiness 101 TechSoup and Remy Consulting
mini mental status format.docx
mini mental status format.docx
Employee wellbeing at the workplace.pptx
Employee wellbeing at the workplace.pptx
Measures of Central Tendency: Mean, Median and Mode
Measures of Central Tendency: Mean, Median and Mode
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
A Critique of the Proposed National Education Policy Reform
A Critique of the Proposed National Education Policy Reform
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
Solving Puzzles Benefits Everyone (English).pptx
Solving Puzzles Benefits Everyone (English).pptx
Introduction to ArtificiaI Intelligence in Higher Education
Introduction to ArtificiaI Intelligence in Higher Education
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
How to Make a Pirate ship Primary Education.pptx
How to Make a Pirate ship Primary Education.pptx
Organic Name Reactions for the students and aspirants of Chemistry12th.pptx
Organic Name Reactions for the students and aspirants of Chemistry12th.pptx
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
_Math 4-Q4 Week 5.pptx Steps in Collecting Data
_Math 4-Q4 Week 5.pptx Steps in Collecting Data
Introduction to AI in Higher Education_draft.pptx
Introduction to AI in Higher Education_draft.pptx
Model Call Girl in Bikash Puri Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Bikash Puri Delhi reach out to us at 🔝9953056974🔝
Micromeritics - Fundamental and Derived Properties of Powders
Micromeritics - Fundamental and Derived Properties of Powders
Hybridoma Technology ( Production , Purification , and Application )
Hybridoma Technology ( Production , Purification , and Application )
Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)
Separation of Lanthanides/ Lanthanides and Actinides
Separation of Lanthanides/ Lanthanides and Actinides
Load models ppt
1.
Creating Effective Load
Models for Performance Testing with Incomplete Empirical Data Updated for the: 3rd World Congress for Software Quality September, 2005 Munich, Germany Initial Research Presented for the: 6th IEEE International Workshop on Web Site Evolution September, 2004 Chicago, IL Scott Barber Chief Technology Officer PerfTestPlus, Inc. www.PerfTestPlus.com Creating Effective Load Models with Page 1 © 2005 PerfTestPlus All rights reserved. Incomplete Empirical Data
2.
Common Modeling Techniques… Tend
to fall into one of two categories: Rigorous – Time consuming – Mathematically intensive and/or complex – High degree of accuracy (when done well) – Requires empirical data Overly Simplistic – Quick – Little to no math needed – Occasionally accurate (generally by accident) – Ignores empirical data There is very little in between to assist the modeler in industry that desires rigor, but barely has time for simplistic! www.PerfTestPlus.com Creating Effective Load Models with Page 2 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
3.
Rigorous Techniques Connie U.
Smith, PhD. - Performance Solutions: A Practical Guide to Creating Responsible, Scalable Software [1] Alberto Savoia - “Web Load Test Planning: Predicting how your Web site will respond to stress" [2] Daniel Menasce, PhD. – Capacity Planning for Web Performance: Metrics, Models and Methods [3] & Scaling for E-Business [4] J.D. Meier - Improving .NET Application Performance and Scalability [5] **All Require Empirical Data www.PerfTestPlus.com Creating Effective Load Models with Page 3 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
4.
Performance Solutions… “Software Performance
Engineering (SPE) Approach” to Building Usage Models: “SPE is a comprehensive way of managing performance that includes principles for creating responsive software, performance patterns and antipatterns for performance- oriented design, techniques for eliciting performance objectives, techniques for gathering the data needed for evaluation, and guidelines for the types of evaluation to be performed at each stage of the development process. “SPE is model-based… By building and analyzing models of the proposed software, we can explore its characteristics to determine if it will meet its requirements before we actually commit to building it”. [6] www.PerfTestPlus.com Creating Effective Load Models with Page 4 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
5.
Performance Solutions… “Performance testing
is vital to confirm that the software’s performance actually meets its performance objective, and that the models are representative of the software’s behavior”. [7] To elaborate on Smith's statement: • Performance testing is vital because models are not perfect. • Validation of the model’s predictions via testing ensures the model matches the implementation. • SPE model predictions are only validated accurate load models. • Chapter 3 details how to model application usage in terms of performance with UML, BUT • There is no discussion about how to determine or estimate actual usage of the application under test. Conclusion: With no empirical (i.e. actual usage) data, SPE models cannot be validated. www.PerfTestPlus.com Creating Effective Load Models with Page 5 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
6.
“Web Load Test
Planning…” “Website Usage Signature (WUS) Approach” to Building Usage Models: “In order to be worth anything, I believe that Web site load tests should reproduce the anticipated loads as accurately and realistically as possible. In order to do that you will need to study previous load patterns and design test scenarios that closely recreate them... ”. [8] Savoia expands, saying: • The preferred method of determining actual load and load patterns is accomplished by analyzing log files from the existing version of the application. • In the absence of relevant log files…[generate] log files via use of a limited beta release of the application to a representative sampling of users Conclusion: With no empirical (i.e. actual usage) data from previous versions or limited releases, WUS is not applicable. www.PerfTestPlus.com Creating Effective Load Models with Page 6 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
7.
Scaling… & Capacity
Planning… “Capacity Planning Approach” to Building Usage Models: In Scaling…[9] Menascé advocates: • WUS-like usage models • SPE-like architectural models • Forecasting mathematically from these In Capacity Planning…[10] Menascé discusses: • Methods for generating models for performance testing and capacity planning when actual usage data isn’t available. • Though these methods are based in mathematical concepts that few practicing performance testers are skilled in. Conclusion: With no empirical (i.e. actual usage) data the methods in Scaling… are not applicable and the mathematics involved with the methods in Capacity Planning… make them unusable in much of industry. www.PerfTestPlus.com Creating Effective Load Models with Page 7 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
8.
Improving .NET… “End-to-End Approach”
to Building Usage Models: “You can determine patterns and call frequency by using either the predicted usage of your application based on market analysis, or if your application is already in production, by analyzing server log files. Some important questions to help you determine the workload for your application include: – What are your key scenarios?... – What is the possible set of actions that a user can perform?... – … – What is the duration for which the test needs to be executed?...” [11] Conclusion: Each question is explained and has examples, but how to collect the data is not addressed, making the method unreliable without empirical data. www.PerfTestPlus.com Creating Effective Load Models with Page 8 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
9.
State-of-the-Practice In Practice:
– Empirical Data is Uncommon – Complex Math Skills are Rare – Time is not a Luxury we have – The Highest Volume is Often on “Go-Live Day” – Few Modeling Tools and Methods are Easily Available – Most Usage Models are little more than “Semi- Educated” Guesses www.PerfTestPlus.com Creating Effective Load Models with Page 9 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
10.
State-of-the-Practice “State-of-the-Practice Approaches” to
Building Usage Models: Common Approaches in Industry: – Evaluation of use cases and design documents. – Information from a key stakeholder (i.e. an interview). – Observe user acceptance and/or usability testing. – Evaluate similar sites and generic web usage data. – Personal use and experience. Which lead to: – Low confidence in predicting performance under actual usage. – High likelihood of undetected performance issues in production. – Little or no validation of the load testing model. – A propensity for accepting load generation tool limitations. www.PerfTestPlus.com Creating Effective Load Models with Page 10 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
11.
Hybrid Approach Based on
the experiences of 13 expert consultants… [12], [13] – Start with available production data from existing versions of the app. – Where there is no empirical usage data, start with interviews and documentation. – Collect data from a limited beta roll-out whenever possible. – Compare/supplement data with generic or competitive data. – Develop and distribute for review one or more draft load models. – Run simple in house usage experiments/surveys. – Enhance the model based on experiments and common sense. – Create best, expected and worst case models based on the distribution of usage data collected. – Develop programs and or scripts to implement the model(s). www.PerfTestPlus.com Creating Effective Load Models with Page 11 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
12.
Hybrid Approach Roughly a
dozen “blind” field tests suggest that… [14],[15],[16],[17] – The SPE, Capacity Planning, WUS, End-to-End and Hybrid approaches: • Are statistically equivalent when followed precisely. • Make predictions with acceptable and explainable degrees of error. • Are relatively easy to enhance to accurately predict production. – State-of-the-Practice approaches: • Do work sometimes, but often yield critically misleading predictions. • Are not enhanced or reused after the initial predictions are made. • As often as not, lead to poor business decisions. – WUS, End-to-End and Hybrid approaches: • Are roughly equivalent in terms of time and difficulty to conduct. • Result in nearly identical predictions for applications with feature additions. • Do not require significant training to implement. – Business Users, Managers, Testers and Developers preferred the pictorial documentation of the Hybrid approach more than 10 to 1 over documentation methods recommended by the other approaches. www.PerfTestPlus.com Creating Effective Load Models with Page 12 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
13.
Summary SPE, Capacity Planning,
WUS, End-to-End and Hybrid approaches to developing usage models typically yield “good enough” predictions. State-of-the-Practice approaches to developing usage models typically yield “unreliable” predictions. SPE and Capacity Planning approaches to developing usage models are rarely applicable in industry. WUS, End-to-End and Hybrid approaches to developing usage models are generally viable typically for industry use. The Hybrid approach is an acceptable and more accurate alternative to State-of-the-Practice approaches to developing usage models when empirical data is not available. www.PerfTestPlus.com Creating Effective Load Models with Page 13 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
14.
References 1.
Smith, Connie U.; Williams, Lloyd G. (2002), Performance Solutions: A Practical Guide to Creating Responsible, Scalable Software, Pearson Education, Inc. 2. Savoia, Alberto (2001) “Web Load Test Planning: Predicting how your Web site will respond to stress”, STQE Magazine. 3. Menascé, Daniel A.; Almeida, Virgilio A.F. (1998), Capacity Planning for Web Performance: Metrics, Models and Methods, Prentice Hall PTR. 4. Menascé, Daniel A.; Almeida, Virgilio A.F. (2000), Scaling for E-Business: Technologies, Models, Performance and Capacity Planning, Prentice Hall PTR. 5. Meier, J.D.; Vasireddy, Srinath; Babbar, Ashish; Mackman, Alex (2004) “Improving .NET Application Performance and Scalability”, http://msdn.microsoft.com/library/default.asp?url=/library/en- us/dnpag/html/scalenet.asp, last accessed June 7, 2004 6. Smith, 2002 7. Smith, 2002 8. Savoia, 2001 9. Menascé, 1988 10. Menascé, 2000 11. Meier, 2004 12. Barber, Scott (2002), “User Experience, not Metrics”, Rational Developer Network 13. Barber, Scott (2003), “Beyond Performance Testing”, Rational Developer Network 14. Tani, Steve, “In search of Data/Case studies #2”, 6/27/2004, via e-mail 15. White, Nathan, “Tool for Overall Workload Distribution Figure”, 9/19/2002, via e-mail 16. Rodgers, Nigel, “UCML”, 6/16/2004, via e-mail 17. Warren, Brian, Ceridian, “UCML 1.1 Visio stencils”, 6/9/2004, via e-mail www.PerfTestPlus.com Creating Effective Load Models with Page 14 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
15.
Questions
www.PerfTestPlus.com Creating Effective Load Models with Page 15 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
16.
Contact Information
Scott Barber Chief Technology Officer PerfTestPlus, Inc E-mail: Web Site: sbarber@perftestplus.com www.PerfTestPlus.com www.PerfTestPlus.com Creating Effective Load Models with Page 16 Incomplete Empirical Data © 2005 PerfTestPlus All rights reserved.
Descargar ahora