Load Analysis and Prediction for Responsive Interactive Applications Peter A. Dinda David R. O’Hallaron Carnegie Mellon University.

Slides:



Advertisements
Similar presentations
Decomposition Method.
Advertisements

Polynomial Curve Fitting BITS C464/BITS F464 Navneet Goyal Department of Computer Science, BITS-Pilani, Pilani Campus, India.
A Prediction-based Real-time Scheduling Advisor Peter A. Dinda Carnegie Mellon University.
An Evaluation of Linear Models for Host Load Prediction Peter A. Dinda David R. O’Hallaron Carnegie Mellon University.
Part II – TIME SERIES ANALYSIS C5 ARIMA (Box-Jenkins) Models
Regression Analysis Module 3. Regression Regression is the attempt to explain the variation in a dependent variable using the variation in independent.
Trace-based Network Bandwidth Analysis and Prediction Yi QIAO 06/10/2002.
1 Reinforcement Learning Introduction & Passive Learning Alan Fern * Based in part on slides by Daniel Weld.
The Case For Prediction-based Best-effort Real-time Peter A. Dinda Bruce Lowekamp Loukas F. Kallivokas David R. O’Hallaron Carnegie Mellon University.
Chair for Communication Technology (ComTec), Faculty of Electrical Engineering / Computer Science Prediction of Context Time Series Stephan Sigg, Sandra.
Host Load Trace Replay Peter A. Dinda Thesis Seminar 11/23/98.
Probabilistic Aggregation in Distributed Networks Ling Huang, Ben Zhao, Anthony Joseph and John Kubiatowicz {hling, ravenben, adj,
A Prediction Service For Remos and QuO Peter A. Dinda CMU SCS.
Dynamic Mapping of Activation Trees Thesis Proposal January 29, 1998 Peter A. Dinda Committee David O’Hallaron (chair) Thomas Gross Peter Steenkiste Jaspal.
Modeling Host Load Peter A. Dinda Thesis Seminar 2/9/98.
1 Quality Objects: Advanced Middleware for Wide Area Distributed Applications Rick Schantz Quality Objects: Advanced Middleware for Large Scale Wide Area.
On the Constancy of Internet Path Properties Yin Zhang, Nick Duffield AT&T Labs Vern Paxson, Scott Shenker ACIRI Internet Measurement Workshop 2001 Presented.
Responsive Interactive Applications by Dynamic Mapping of Activation Trees February 20, 1998 Peter A. Dinda School of Computer.
Understanding and Predicting Host Load Peter A. Dinda Carnegie Mellon University
Resource Signal Prediction and Its Application to Real-time Scheduling Advisors or How to Tame Variability in Distributed Systems Peter A. Dinda Carnegie.
LAB 3 AIRBAG DEPLOYMENT SENSOR PREDICTION NETWORK Warning This lab could save someone’s life!
Recent Results in Resource Signal Measurement, Dissemination, and Prediction App Transport Network Data Link Physical App Transport Network Data Link Physical.
A Prediction-based Approach to Distributed Interactive Applications Peter A. Dinda Jason Skicewicz Dong Lu Prescience Lab Department of Computer Science.
The Measured Network Traffic of Compiler Parallelized Programs Peter A. Dinda Northwestern University, Computer Science
ARIMA Forecasting Lecture 7 and 8 - March 14-16, 2011
Online Prediction of the Running Time Of Tasks Peter A. Dinda Department of Computer Science Northwestern University
+ Doing More with Less : Student Modeling and Performance Prediction with Reduced Content Models Yun Huang, University of Pittsburgh Yanbo Xu, Carnegie.
From Controlled to Natural Settings
A Prediction-based Approach to Distributed Interactive Applications Peter A. Dinda Department of Computer Science Northwestern University
A Prediction-based Real-time Scheduling Advisor Peter A. Dinda Prescience Lab Department of Computer Science Northwestern University
Inferring the Topology and Traffic Load of Parallel Programs in a VM environment Ashish Gupta Peter Dinda Department of Computer Science Northwestern University.
The Running Time Advisor A Resource Signal-based Approach to Predicting Task Running Time and Its Applications Peter A. Dinda Carnegie Mellon University.
Realistic CPU Workloads Through Host Load Trace Playback Peter A. Dinda David R. O’Hallaron Carnegie Mellon University.
G. Peter Zhang Neurocomputing 50 (2003) 159–175 link Time series forecasting using a hybrid ARIMA and neural network model Presented by Trent Goughnour.
Host Load Prediction in a Google Compute Cloud with a Bayesian Model Sheng Di 1, Derrick Kondo 1, Walfredo Cirne 2 1 INRIA 2 Google.
Traffic modeling and Prediction ----Linear Models
Towards Network Containment in Malware Analysis Systems Authors: Mariano Graziano, Corrado Leita, Davide Balzarotti Source: Annual Computer Security Applications.
1 Patch Complexity, Finite Pixel Correlations and Optimal Denoising Anat Levin, Boaz Nadler, Fredo Durand and Bill Freeman Weizmann Institute, MIT CSAIL.
1 A User-Guided Cognitive Agent for Wireless Service Selection in Pervasive Computing George Lee May 5, 2004 G. Lee, P. Faratin, S. Bauer, and J. Wroclawski.
Master Thesis Defense Jan Fiedler 04/17/98
Fan Guo 1, Chao Liu 2 and Yi-Min Wang 2 1 Carnegie Mellon University 2 Microsoft Research Feb 11, 2009.
FRE 2672 TFG Self-Organization - 01/07/2004 Engineering Self-Organization in MAS Complex adaptive systems using situated MAS Salima Hassas LIRIS-CNRS Lyon.
June 05 David A. Gaitros Jean Muhammad Introduction to OOD and UML Dr. Jean Muhammad.
MURI: Integrated Fusion, Performance Prediction, and Sensor Management for Automatic Target Exploitation 1 Dynamic Sensor Resource Management for ATE MURI.
Maximum Likelihood Estimation Methods of Economic Investigation Lecture 17.
1 Distribution Statement “A” (Approved for Public Release, Distribution Unlimited)5/15/2012 Advanced Radio Frequency Mapping (RadioMap) Dr. John Chapin.
FORECASTING. Minimum Mean Square Error Forecasting.
Big Data at Home Depot KSU – Big Data Survey Course Steve Einbender Advanced Analytics Architect.
Chapter 10 Verification and Validation of Simulation Models
Experiments on Noise CharacterizationRoma, March 10,1999Andrea Viceré Experiments on Noise Analysis l Need of noise characterization for  Monitoring the.
Integration of QoS-enabled Distributed Object Computing Middleware for Developing Next- Generation Distributed Applications By Krishnamurthy et Al. Presented.
How Good is a Model? How much information does AIC give us? –Model 1: 3124 –Model 2: 2932 –Model 3: 2968 –Model 4: 3204 –Model 5: 5436.
NTU & MSRA Ming-Feng Tsai
ETM 607 – Putting It All Together Review for MidTerm II Apply Lessons Learned in a Team Lab - Input Modeling - Absolute Output Analysis - Relative Output.
A Brief Maximum Entropy Tutorial Presenter: Davidson Date: 2009/02/04 Original Author: Adam Berger, 1996/07/05
Irwin/McGraw-Hill © Andrew F. Siegel, 1997 and l Chapter 14 l Time Series: Understanding Changes over Time.
Data Mining: Neural Network Applications by Louise Francis CAS Convention, Nov 13, 2001 Francis Analytics and Actuarial Data Mining, Inc.
Subodh Kant. Auto-Regressive Integrated Moving Average Also known as Box-Jenkins methodology A type of linear model Capable of representing stationary.
Accurate WiFi Packet Delivery Rate Estimation and Applications Owais Khan and Lili Qiu. The University of Texas at Austin 1 Infocom 2016, San Francisco.
Statistics 350 Lecture 2. Today Last Day: Section Today: Section 1.6 Homework #1: Chapter 1 Problems (page 33-38): 2, 5, 6, 7, 22, 26, 33, 34,
Career Related Applications of Economics In Enrollment Management Matt Bogard Coordinator, Market Research Western Kentucky University.
Forecasting. Model with indicator variables The choice of a forecasting technique depends on the components identified in the time series. The techniques.
Lecture 9 Forecasting. Introduction to Forecasting * * * * * * * * o o o o o o o o Model 1Model 2 Which model performs better? There are many forecasting.
Operations Management Contemporary Concepts and Cases
Forecasting Chapter 11.
Partially Predictable
Chapter 10 Verification and Validation of Simulation Models
S519: Evaluation of Information Systems
Model diagnostics Tim Paine, modified from Zarah Pattison’s slides
Partially Predictable
Presentation transcript:

Load Analysis and Prediction for Responsive Interactive Applications Peter A. Dinda David R. O’Hallaron Carnegie Mellon University

2 Overview Responsive Interactive Applications (eg, BBN OpenMap) Best Effort Real-time Communication Execution Time Predicition Computation History-based Load Prediction Load AnalysisTime Series Modelling Remote Execution Measurement

3 OpenMap (BBN) “Move North” New map data Integrator Choice of Host Terrain Bounded Response Time Replicated Specialists

4 Context OpenMap (BBN) Load Prediction (CMU) QuO (BBN) Remos (CMU) JTF Planner Advanced Mobility Platform Logistics Anchor Desk METOC Anchor Desk TRACE2ES... Other Applications Frameworks Adaptation Measurement Prediction Applications Distributed system

5 Statistical Analysis Appropriate Time Series Models Fitted Models Evaluation/ Comparison On-line Predictors Load Trace Collection Load Analysis and Prediction Goal: accurate short term predictions –Few seconds for non-stale data Evaluation/comparison issues –Load generation vs. Load prediction Have to discover which properties are important –Performance measure Mean squared prediction error Lack of lower bound to compare against Simple, reasonable algorithm for comparison

6 Load Trace Analysis Digital Unix one minute load average Four classes of hosts (38 machines) 1 Hz sample rate, >one week traces, two sets at different times of the year Analysis results to appear in LCR98 Load is self-similar Load exhibits epochal behavior

7 Self-similarity Statistics

8 Why is Self-Similarity Important? Complex structure –Not completely random, nor independent –Short range dependence Excellent for history-based prediction –Long range dependence Possibly a problem Modeling Implications –Suggests models ARFIMA, FGN, TAR

9 Load Exhibits Epochal Behavior

10 Epoch Length Statistics

11 Why is Epochal Behavior Important? Complex structure –Non-stationary Modeling Implications –Suggests models ARIMA, ARFIMA, etc. Non-parametric spectral methods –Suggests problem decomposition

12 Time Series Prediction of Load Linear Nonlinear StationaryNon-stationary ARMA, AR, MA ARIMAARFIMA, FGN TARMarkov Self-similarNon-self-similar “Best Mean” Non-parametricParametric

13 t+1 Predictions

14 t+5 Prediction

15 Conclusions Load has structure to exploit for prediction Structure is complex (self-similarity, epochs) Simple time series models are promising Benefits of more sophisticated models are unclear Current research questions What are the benefits of more sophisticated models? How to characterize prediction error to user? Is there a measure of inherent predictability? How to incorporate load prediction into systems?

16