IEEE 7th Annual Workshop on Workload Characterization The USAR Characterization Model Adriano Pereira, Gustavo Gorgulho, Leonardo Silva, Wagner Meira Jr.,

Slides:



Advertisements
Similar presentations
Andrea Maurino Web Service Design Methodology Batini, De Paoli, Maurino, Grega, Comerio WP2-WP3 Roma 24/11/2005.
Advertisements

Inktomi Confidential and Proprietary The Inktomi Climate Lab: An Integrated Environment for Analyzing and Simulating Customer Network Traffic Stephane.
Autonomic Scaling of Cloud Computing Resources
Han-na Yang Trace Clustering in Process Mining M. Song, C.W. Gunther, and W.M.P. van der Aalst.
Interception of User’s Interests on the Web Michal Barla Supervisor: prof. Mária Bieliková.
Experiments on Query Expansion for Internet Yellow Page Services Using Log Mining Summarized by Dongmin Shin Presented by Dongmin Shin User Log Analysis.
Introduction to Research Methodology
Computer Science Generating Streaming Access Workload for Performance Evaluation Shudong Jin 3nd Year Ph.D. Student (Advisor: Azer Bestavros)
Hiperspace Lab University of Delaware Antony, Sara, Mike, Ben, Dave, Sreedevi, Emily, and Lori.
Set-Based Model: A New Approach for Information Retrieval Bruno Pôssas Nivio Ziviani Wagner Meira Jr. Berthier Ribeiro-Neto Department of Computer Science.
Project 4 U-Pick – A Project of Your Own Design Proposal Due: April 14 th (earlier ok) Project Due: April 25 th.
October 14, 2002MASCOTS Workload Characterization in Web Caching Hierarchies Guangwei Bai Carey Williamson Department of Computer Science University.
1 Experimental Methodology H Experimental methods can be used to: – demonstrate that a new concept, technique, or algorithm is feasible –demonstrate that.
1 Internet Protocols and Network Performance Issues Carey Williamson iCORE Professor Department of Computer Science University of Calgary.
1 Simulation Evaluation of a Heterogeneous Web Proxy Caching Hierarchy Mudashiru Busari Carey Williamson University of Saskatchewan University of Calgary.
A Hierarchical Characterization of a Live Streaming Media Workload E. Veloso, V. Almeida W. Meira, A. Bestavros, S. Jin Proceedings of Internet Measurement.
OS Fall ’ 02 Performance Evaluation Operating Systems Fall 2002.
Looking at the Server-side of P2P Systems Yi Qiao, Dong Lu, Fabian E. Bustamante and Peter A. Dinda Department of Computer Science Northwestern University.
A Hierarchical Characterization of a Live Streaming Media Workload IEEE/ACM Trans. Networking, Feb Eveline Veloso, Virg í lio Almeida, Wagner Meira,
Web Mining Research: A Survey
WebMiningResearchASurvey Web Mining Research: A Survey Raymond Kosala and Hendrik Blockeel ACM SIGKDD, July 2000 Presented by Shan Huang, 4/24/2007 Revised.
Internet Cache Pollution Attacks and Countermeasures Yan Gao, Leiwen Deng, Aleksandar Kuzmanovic, and Yan Chen Electrical Engineering and Computer Science.
A Survey of proxy Cache Evaluation Techniques 系統實驗室 田坤銘
Efficiently Maintaining Stock Portfolios Up-To-Date On The Web Prashant Shenoy Manish Bhide Krithi Ramamritham 2002 IEEE E-Commerce System Proceedings.
LDU Parametrized Discrete-Time Multivariable MRAC and Application to A Web Cache System Ying Lu, Gang Tao and Tarek Abdelzaher University of Virginia.
OS Fall ’ 02 Performance Evaluation Operating Systems Fall 2002.
SIMULATION. Simulation Definition of Simulation Simulation Methodology Proposing a New Experiment Considerations When Using Computer Models Types of Simulations.
Sujit Dey Adaptive Applications for Wireless Information Technology Sujit Dey ECE Department University of California, San Diego
Personalized Ontologies for Web Search and Caching Susan Gauch Information and Telecommunications Technology Center Electrical Engineering and Computer.
IE 594 : Research Methodology – Discrete Event Simulation David S. Kim Spring 2009.
Recommender Systems on the Web: A Model-Driven Approach Gonzalo Rojas – Francisco Domínguez – Stefano Salvatori Department of Computer Science University.
Analyzing Seller Practices in a Brazilian Marketplace Adriano Pereira Diego Duarte Wagner Meira Jr.
Performance of Web Applications Introduction One of the success-critical quality characteristics of Web applications is system performance. What.
1 An SLA-Oriented Capacity Planning Tool for Streaming Media Services Lucy Cherkasova, Wenting Tang, and Sharad Singhal HPLabs,USA.
Dr. Russell Anderson Dr. Musa Jafar West Texas A&M University.
 1  Outline  stages and topics in simulation  generation of random variates.
Modeling and Performance Evaluation of Network and Computer Systems Introduction (Chapters 1 and 2) 10/4/2015H.Malekinezhad1.
Web Cache Replacement Policies: Properties, Limitations and Implications Fabrício Benevenuto, Fernando Duarte, Virgílio Almeida, Jussara Almeida Computer.
Workload-driven Analysis of File Systems in Shared Multi-Tier Data-Centers over InfiniBand K. Vaidyanathan P. Balaji H. –W. Jin D.K. Panda Network-Based.
nd Joint Workshop between Security Research Labs in JAPAN and KOREA Profile-based Web Application Security System Kyungtae Kim High Performance.
Information System Development Courses Figure: ISD Course Structure.
Quantitative Evaluation of Unstructured Peer-to-Peer Architectures Fabrício Benevenuto José Ismael Jr. Jussara M. Almeida Department of Computer Science.
1 Challenges in Scaling E-Business Sites  Menascé and Almeida. All Rights Reserved. Daniel A. Menascé Department of Computer Science George Mason.
Simulation is the process of studying the behavior of a real system by using a model that replicates the behavior of the system under different scenarios.
Performance evaluation of component-based software systems Seminar of Component Engineering course Rofideh hadighi 7 Jan 2010.
1 Evaluation of Cooperative Web Caching with Web Polygraph Ping Du and Jaspal Subhlok Department of Computer Science University of Houston presented at.
McGraw-Hill/Irwin © 2006 The McGraw-Hill Companies, Inc., All Rights Reserved. 1.
Carnegie Mellon University Computer Science Department 1 OPEN VERSUS CLOSED: A CAUTIONARY TALE Bianca Schroeder Adam Wierman Mor Harchol-Balter Computer.
IAT 814 Introduction to Visual Analytics Symbols vs Perceptual Science Sep 11, 2013IAT 8141.
DEVS Based Modeling and Simulation of the CORBA POA F. Bernardi, E. de Gentili, Pr. J.F. Santucci {bernardi, gentili, University.
Unconstrained Endpoint Profiling Googling the Internet Ionut Trestian, Supranamaya Ranjan, Alekandar Kuzmanovic, Antonio Nucci Reviewed by Lee Young Soo.
© 2006 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice Injecting Realistic Burstiness to.
McGraw-Hill/Irwin Copyright © 2009 by The McGraw-Hill Companies, Inc. All rights reserved.
Stefanos Antaris A Socio-Aware Decentralized Topology Construction Protocol Stefanos Antaris *, Despina Stasi *, Mikael Högqvist † George Pallis *, Marios.
OPERATING SYSTEMS CS 3530 Summer 2014 Systems and Models Chapter 03.
Chaoyang University of Technology Clustering web transactions using rough approximation Source : Fuzzy Sets and Systems 148 (2004) 131–138 Author : Supriya.
The basics of knowing the difference CLIENT VS. SERVER.
Predicting the Location and Time of Mobile Phone Users by Using Sequential Pattern Mining Techniques Mert Özer, Ilkcan Keles, Ismail Hakki Toroslu, Pinar.
Identifying “Best Bet” Web Search Results by Mining Past User Behavior Author: Eugene Agichtein, Zijian Zheng (Microsoft Research) Source: KDD2006 Reporter:
On the Benefits of Planning and Grouping Software Maintenance Requests CSMR – Oldenburg, Germany, March 2011 Gladston Aparecido Junio (PUC Minas, Brazil)
Distinguishing humans from robots in web search logs preliminary results using query rates and intervals Omer Duskin Dror G. Feitelson School of Computer.
M. R. Gouvea BR Session 1 – Block 1 – Transformers – Paper #36 Barcelona May RISK CRITERIA TO APPLY AND MANAGE DISTRIBUTION TRANSFORMERS M.
1 Evaluation of Cooperative Web Caching with Web Polygraph Ping Du and Jaspal Subhlok Department of Computer Science University of Houston presented at.
LIOProf: Exposing Lustre File System Behavior for I/O Middleware
A Practical Performance Analysis of Stream Reuse Techniques in Peer-to-Peer VoD Systems Leonardo B. Pinho and Claudio L. Amorim Parallel Computing Laboratory.
Lin Lu, Margaret Dunham, and Yu Meng
A Unifying View on Instance Selection
Computer Systems Performance Evaluation
Assignment of Games to Servers in the OnLive Cloud Game System
Computer Systems Performance Evaluation
Presentation transcript:

IEEE 7th Annual Workshop on Workload Characterization The USAR Characterization Model Adriano Pereira, Gustavo Gorgulho, Leonardo Silva, Wagner Meira Jr., and Walter Santos Department of Computer Science Federal University of Minas Gerais (UFMG) Belo Horizonte - Brazil October, 2004

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Presentation Index Introduction Goals Methodology –Characterization Model –Validation Model Case Study Conclusion and Ongoing Work

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Introduction Understanding the nature and characteristics of Web workloads is a crucial step to improve the quality of the offered service –Design systems with better Performance and Scalability. Workload Characterization Methodologies and Techniques –Guidelines to characterize and generate workloads

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Introduction Workload Characterizations typically do not consider reactivity aspects –Do not consider agreggate information from user and server-side –Generate the same workload despite the server response time Server Client Request Response

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Goals Characterize and replicate the behavior related to user reactivity Analyze and model the way users react to variations in the quality of service provided. –Correlate user and server-side Validate the model through simulation.

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Methodology When we look at the interaction process we have sequences of requests and responses –IAT (inter-arrival time): time between requests User-side related measure –Latency: time to process and answer the request Server-side related measure LATENCY IAT Server Client REQUEST RESPONSE

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Methodology The USAR Characterization Model –Conceptual views USER (U) SESSION (S) ACTION (A) REQUEST (R) User: user behavior considering quality of service Session: actions between a threshold τ Action: clicks from the user Requests: objects associated with a action

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory User Level Characterization 1) Prepare log: generate a temporary log Lu by aggregating the sessions per user; 2) Analyze users from the following perspectives: –IAT and Latency ratio; –IAT and Latency difference; 3) Discretize IAT and Latency measures using a correlation function;

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory User Level Characterization - Discretization Model 7 User Classes (A – G) DIF k5 k6 RAT C B A E F G D 0k1k2 k3 k4 PatienceImpatience

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory User Level Characterization 4) Transform user sessions into sequences of user classes according to discretization criteria from item 3); –Ex: A A B F B G G G G A 5) Evaluate the sequences in order to group them using similarity; –Sequence mining (SPADE tool) or matching

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory User Level Characterization Classify them in User Profiles: – –Patientk5 k6 RAT C B A E F G D 0k1k2 k3 k4 – –Impatient – –Impatient Tendency – –Continuous – –Patient Tendency – –Inconstant

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory User Level Characterization 6) Process the log Lu applying a function f(Lu), which maps sequence of classes to groups defined in step 5; 7) Apply Clustering; 8) Analyze the clusters and classify them.

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Methodology Validation using the USAR Simulation Model USARUSAR Inputs Users Types Behaviors and Profiles Session Length Dist. Action Popularity Object Size Object Popularity Parameters # User Sessions Log of User Reactions per Session

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Case Study Proxy-cache of Federal University of Minas Gerais (UFMG); 4 weeks of logs.

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Case Study Distribution of Sessions and User Profiles

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Case Study Validation through simulation USAR

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Conclusion We propose the USAR characterization methodology that comprehend the levels of request, action, session and user –Generic Methodology applicable to any workload Model reactivity aspects using inter-arrival time and latency that has been interesting and promising –Present a discretization model based on the radio and difference between IAT and latency Validate the model through simulation

e-Commerce, Systems Performance Evaluation, and Experimental Development Laboratory Ongoing Work We foresee the possibility to generate more realistic workload –Reduce the gap between the existing models and the actual workloads considering reactivity aspects