SlideShare a Scribd company logo
Proactive Prediction Models for Web
Application Resource Provisioning in the Cloud
_______________________________
Samuel A. Ajila & Bankole A. Akindele
Presentation Outline
 Introduction to Problem Area
 Motivation
 Goals and Scope
 Contributions
 Related work
 Machine learning algorithms
 Implementation setup
 Evaluation metrics
 Selected results
 Conclusion
2
Introduction
 Cloud computing refers to both the applications delivered
as services over the Internet and the hardware and
systems software in the data centers that provide those
services (SaaS, PaaS and IaaS)
 Challenges include: Data security threats, performance
unpredictability and prompt (quick) resource scaling
 Accurate Virtual machine (VM) resource provisioning:
• under and over-provisioning
 Current techniques like Control theory, Constraint
programming and Machine learning
3
Motivation
 VM instantiation (scaling) takes time and ranges from 5 to 12
minutes
 Challenges that can result from the instantiation duration:
• Possibility of Service Level Agreement (SLA) violations -
Cloud providers
• Poor customer’s Quality of Experience (QoE) – Cloud client
• Reputation loss – Cloud client/providers
 Presently, monitoring metrics made available to clients are
limited to CPU, Memory and Network utilization
 The motivation is to “predict resource usage so that cloud
providers can make adequate provisioning ahead of time”
 To extend the monitoring metrics by including Response time
and Throughput 4
Goals and Scope
 Design and develop a cloud client prediction model for cloud
resource provisioning in a Multitier web application
environment
 The model would be capable of forecasting future resource
usage to enable timely VM provisioning
 To achieve this goal, SVM, NN and LR learning techniques are
analysed using the Java implementation of TPC-W (workload)
 The scope of this work is limited to IaaS
 Prediction model is built around the web server tier. It is
possible to extend to other tiers
5
Contributions
 Design and development of a cloud client prediction model that
uses historical data to forecast future resource usage
 The evaluation of the resource usage prediction capability of
SVM, NN and LR using three benchmark workloads from TPC-W
 The extension of the prediction model to include Throughput
and Response time, thus providing wider and better scaling
decision options for cloud clients
 The comparison of the prediction capability of SVM, NN and LR
models under random and steady traffic patterns
6
Related works
Table 1: Auto-scaling techniques
7
Technique Class Comments
Threshold/Rule
based
Reactive Reacts to system changes but do not
anticipate them
Control theory Reactive/Proactive Except where used with a proactive
approach it suffers reactive issues
Queuing theory Reactive As complexity of system grows, the
analytic formulas become difficult
Reinforcement
learning
Reactive/Proactive Good but converging at optimal
policy can be unfeasibly long (state
–action pairs)
Time series analysis Proactive Good and promising
Machine learning algorithms
Linear Regression
 The linear regression model has the form: 𝑓 𝑋 = 𝛽0 + 𝑗=1
𝑝
𝑋𝑗 𝛽𝑗 (1)
 𝑋𝑗 is the input data
 𝛽0 and 𝛽𝑗’s are the unknown parameters estimated from the training
data. Estimation is done using the least squares method.
 The coefficients 𝛽 are picked to minimize the residual (difference
between the actual and predicted value) sum of squares :
(𝑦𝑖 − 𝑓 𝑋 )2
𝑦𝑖 is the actual value
8
Machine learning algorithms (cont’d)
Neural Network
 A two-stage regression or
classification model represented
by a network diagram
 Derived features 𝑍 𝑚 are created
from linear combinations of the
input after which the target 𝑌𝑘 is
modeled as a linear combination
of 𝑍 𝑚
 Like Linear regression, unknown
parameters called weights are
sought to make the model fit the
training data well
9
Figure 1 : Single hidden layer, feed forward neural network
Machine learning algorithms (cont’d)
Support Vector Regression
 The goal of SVR is to find a function 𝑓 𝑋 that has at most 𝜀 (the
precision by which the function is to be approximated) deviation from
the actual obtained target for all training data
 Mathematically, 𝑓 𝑋 = 𝑤. ∅ 𝑥 + 𝑏
 Input data 𝑥 is mapped to higher dimensional feature space via the
kernel function, then a linear regression is performed
 Goal is to find optimal weights 𝑤 and threshold 𝑏
10
Architecture
11Figure 2 Implementation architecture
Implementation setup (cont’d)
Time (minutes) 1-7 56-63 154-161 350-357 490-497 504-511
Shopping mix
users
84 168 16 180 248 160
Browsing mix
users
52 112 36 320 192 160
Ordering mix
users
52 108 28 224 268 160
Total user
Requests
188 388 80 724 708 480
12
 Total length of experiment was about 10 hours
 Selected experimental workload mix
Evaluation metrics
13
 Evaluation is done on the 60% training and 40% held out test
dataset
 The held out dataset is used to forecast to a maximum interval
of 12 minutes (VM instantiation time reported by other
authors)
Performance metrics and their calculations
Metric Calculation
MAPE (Mean Absolute
Percentage Error)
1
𝑛
| 𝑎𝑖−𝑝𝑖|
𝑎𝑖
𝑛
𝑖=1 where 𝑎𝑖 and 𝑝𝑖 are the actual and predicted values
respectively
RMSE (Root Mean Square
Error)
√
( 𝑎𝑖−𝑝𝑖) 2𝑛
𝑖=1
𝑛
MAE (Mean Absolute Error) 1
𝑛
|𝑝𝑖 − 𝑎𝑖|𝑛
𝑖=1
PRED 25 No. of observations with relative error ≤ 25% / No. of observation
CPU Training & Testing Performance
14
CPUUtilizationTrainingandTestingPerformanceMetricforLR,NN,andSVR
LR(LinearRegression)Model MAPE RMSE MAE PRED(25)
LRTraining 113.31 14.70 11.11 0.51
LRTesting 36.19 22.13 15.98 0.36
NN(NeuralNetwork)Model MAPE RMSE MAE PRED(25)
NNTraining 105.63 14.08 9.48 0.59
NNTesting 50.46 31.08 19.82 0.34
SVR(SupportVectorRegression)Model MAPE RMSE MAE PRED(25)
SVRTraining 107.80 15.48 10.09 0.64
SVRTesting 22.84 11.84 8.74 0.64
Selected results (cont’d)
15
CPU utilization test performance metric
Model MAPE RMSE MAE PRED(25)
LR 36.19 22.13 15.98 0.36
NN 50.46 31.08 19.82 0.34
SVR 22.84 11.84 8.74 0.64
Figure 5 CPU Utilization Actual and Predicted test model results
Selected results (cont’d)
16
Throughput test performance metric
Model MAPE RMSE MAE PRED(25)
LR 24.62 3.72 2.87 0.63
NN 38.90 6.12 4.46 0.47
SVR 22.07 3.22 2.41 0.67
Figure 6 Throughput Actual and Predicted test model results
Selected results (cont’d)
17
Response time test performance metric
Model MAPE RMSE MAE PRED(25)
LR 12.35 1.39 1.11 0.91
NN 17.84 2.02 1.64 0.75
SVR 9.92 1.21 0.87 0.93
Figure 7 Response time Actual and Predicted test model results
CPU - Comparison of Prediction Models
18
Throughput – Comparison of Prediction
Models
19
Sensitivity Measurement Using Little’s Law
20
Data Consistency Measurement Using Little’sLaw
Time (minute) 1-7 56-63 154-
161
350-
357
490-
497
498-
503
504-
511
Average total user
requests
27
(188/7)
55
(388/7)
11
(80/7)
103
(724/7)
101
(708/7)
95
(664/7)
69
(480/7)
Average Throughput
(Requests/second)
5.29 10.52 2.47 15.92 10.00 12.23 13.27
Average time spent
(seconds)
5.10 5.23 4.45 6.47 10.01 7.77 5.2
Measured time spent 4.66 4.92 9.92 7.53 9.06 9.01 9.45
Time variance (%) 9.44 6.31 55.14 14.08 10.49 13.76 44.97
Conclusion
 SVR displayed superior prediction accuracy over both LR and
NN in a typically nonlinear, not defined a-priori workload by:
 In the CPU utilization prediction model, SVR
outperformed LR and NN by 58% and 120% respectively
 For the Throughput prediction model, SVR again
outperformed LR and NN by 12% and 76% respectively;
and finally,
 The Response time prediction model saw SVR
outperforming LR and NN by 26% and 80% respectively.
 Based on this experimental results SVR may be accepted as
the best prediction model in a nonlinear, not defined a-priori
system
21
Future works
 SVR and other machine learning algorithms are good for
forecasting, however
• training and retraining is a challenge
• Parameter selection is still empirical
 Combination of SVR and other predicting techniques may
mitigate this challenge
 Other future direction include
• Inclusion of database tier for a more robust scaling
decision
• Resource prediction on other non-web application
workload
22
Questions
23
Thank You

More Related Content

What's hot

Implementation of area optimized low power multiplication and accumulation
Implementation of area optimized low power multiplication and accumulationImplementation of area optimized low power multiplication and accumulation
Implementation of area optimized low power multiplication and accumulation
karthik annam
 
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions   2014 ieee java project - a hyper-heuristic schedulingIeeepro techno solutions   2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic scheduling
hemanthbbc
 
A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...
eSAT Journals
 
A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...
eSAT Publishing House
 
IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...
IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...
IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...
IRJET Journal
 
RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...
RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...
RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...
ijdms
 
Iaetsd improved load balancing model based on
Iaetsd improved load balancing model based onIaetsd improved load balancing model based on
Iaetsd improved load balancing model based on
Iaetsd Iaetsd
 
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
csandit
 
Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...
Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...
Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...
IRJET Journal
 
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...
AtakanAral
 
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]
AtakanAral
 
Learning Software Performance Models for Dynamic and Uncertain Environments
Learning Software Performance Models for Dynamic and Uncertain EnvironmentsLearning Software Performance Models for Dynamic and Uncertain Environments
Learning Software Performance Models for Dynamic and Uncertain Environments
Pooyan Jamshidi
 
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
IDES Editor
 
IMQA Paper
IMQA PaperIMQA Paper
IMQA Paper
Vignesh Kannan
 
Cost Efficient PageRank Computation using GPU : NOTES
Cost Efficient PageRank Computation using GPU : NOTESCost Efficient PageRank Computation using GPU : NOTES
Cost Efficient PageRank Computation using GPU : NOTES
Subhajit Sahu
 
Energy efficient-resource-allocation-in-distributed-computing-systems
Energy efficient-resource-allocation-in-distributed-computing-systemsEnergy efficient-resource-allocation-in-distributed-computing-systems
Energy efficient-resource-allocation-in-distributed-computing-systems
Cemal Ardil
 
Multi PPT - Agent Actor-Critic for Mixed Cooperative-Competitive Environments
Multi PPT - Agent Actor-Critic for Mixed Cooperative-Competitive EnvironmentsMulti PPT - Agent Actor-Critic for Mixed Cooperative-Competitive Environments
Multi PPT - Agent Actor-Critic for Mixed Cooperative-Competitive Environments
Jisang Yoon
 
A survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmentA survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environment
eSAT Publishing House
 
E01113138
E01113138E01113138
E01113138
IOSR Journals
 

What's hot (19)

Implementation of area optimized low power multiplication and accumulation
Implementation of area optimized low power multiplication and accumulationImplementation of area optimized low power multiplication and accumulation
Implementation of area optimized low power multiplication and accumulation
 
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions   2014 ieee java project - a hyper-heuristic schedulingIeeepro techno solutions   2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic scheduling
 
A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...
 
A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...A vm scheduling algorithm for reducing power consumption of a virtual machine...
A vm scheduling algorithm for reducing power consumption of a virtual machine...
 
IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...
IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...
IRJET- Advance Approach for Load Balancing in Cloud Computing using (HMSO) Hy...
 
RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...
RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...
RE-OPTIMIZATION FOR MULTI-OBJECTIVE CLOUD DATABASE QUERY PROCESSING USING MAC...
 
Iaetsd improved load balancing model based on
Iaetsd improved load balancing model based onIaetsd improved load balancing model based on
Iaetsd improved load balancing model based on
 
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
 
Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...
Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...
Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) A...
 
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Progres...
 
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]
Modeling and Optimization of Resource Allocation in Cloud [PhD Thesis Proposal]
 
Learning Software Performance Models for Dynamic and Uncertain Environments
Learning Software Performance Models for Dynamic and Uncertain EnvironmentsLearning Software Performance Models for Dynamic and Uncertain Environments
Learning Software Performance Models for Dynamic and Uncertain Environments
 
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
 
IMQA Paper
IMQA PaperIMQA Paper
IMQA Paper
 
Cost Efficient PageRank Computation using GPU : NOTES
Cost Efficient PageRank Computation using GPU : NOTESCost Efficient PageRank Computation using GPU : NOTES
Cost Efficient PageRank Computation using GPU : NOTES
 
Energy efficient-resource-allocation-in-distributed-computing-systems
Energy efficient-resource-allocation-in-distributed-computing-systemsEnergy efficient-resource-allocation-in-distributed-computing-systems
Energy efficient-resource-allocation-in-distributed-computing-systems
 
Multi PPT - Agent Actor-Critic for Mixed Cooperative-Competitive Environments
Multi PPT - Agent Actor-Critic for Mixed Cooperative-Competitive EnvironmentsMulti PPT - Agent Actor-Critic for Mixed Cooperative-Competitive Environments
Multi PPT - Agent Actor-Critic for Mixed Cooperative-Competitive Environments
 
A survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmentA survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environment
 
E01113138
E01113138E01113138
E01113138
 

Similar to Ajila (1)

A scalable collaborative filtering framework based on co-clustering
A scalable collaborative filtering framework based on co-clusteringA scalable collaborative filtering framework based on co-clustering
A scalable collaborative filtering framework based on co-clustering
lau
 
Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013
Pedro Lopes
 
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
csandit
 
An Adaptive Load Balancing Middleware for Distributed Simulation
An Adaptive Load Balancing Middleware for Distributed SimulationAn Adaptive Load Balancing Middleware for Distributed Simulation
An Adaptive Load Balancing Middleware for Distributed Simulation
Gabriele D'Angelo
 
A Hierarchical Feature Set optimization for effective code change based Defec...
A Hierarchical Feature Set optimization for effective code change based Defec...A Hierarchical Feature Set optimization for effective code change based Defec...
A Hierarchical Feature Set optimization for effective code change based Defec...
IOSR Journals
 
Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)
Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)
Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)
Eswar Publications
 
Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)
Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)
Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)
Peter Tröger
 
IRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware PerformanceIRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware Performance
IRJET Journal
 
IRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor DriveIRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET Journal
 
Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...
Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...
Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...
IJECEIAES
 
Thesis Presentation on Energy Efficiency Improvement in Data Centers
Thesis Presentation on Energy Efficiency Improvement in Data CentersThesis Presentation on Energy Efficiency Improvement in Data Centers
Thesis Presentation on Energy Efficiency Improvement in Data Centers
Monica Vitali
 
Svm on cloud (presntation)
Svm on cloud  (presntation)Svm on cloud  (presntation)
Svm on cloud (presntation)
Ghazanfar Latif (Gabe)
 
Svm Classifier Algorithm for Data Stream Mining Using Hive and R
Svm Classifier Algorithm for Data Stream Mining Using Hive and RSvm Classifier Algorithm for Data Stream Mining Using Hive and R
Svm Classifier Algorithm for Data Stream Mining Using Hive and R
IRJET Journal
 
Maestro_Abstract
Maestro_AbstractMaestro_Abstract
Maestro_Abstract
Hardik Patel
 
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
Tahmid Abtahi
 
Configuration Navigation Analysis Model for Regression Test Case Prioritization
Configuration Navigation Analysis Model for Regression Test Case PrioritizationConfiguration Navigation Analysis Model for Regression Test Case Prioritization
Configuration Navigation Analysis Model for Regression Test Case Prioritization
ijsrd.com
 
M3AT: Monitoring Agents Assignment Model for the Data-Intensive Applications
M3AT: Monitoring Agents Assignment Model for the Data-Intensive ApplicationsM3AT: Monitoring Agents Assignment Model for the Data-Intensive Applications
M3AT: Monitoring Agents Assignment Model for the Data-Intensive Applications
VladislavKashansky
 
ESTIMATING HANDLING TIME OF SOFTWARE DEFECTS
ESTIMATING HANDLING TIME OF SOFTWARE DEFECTSESTIMATING HANDLING TIME OF SOFTWARE DEFECTS
ESTIMATING HANDLING TIME OF SOFTWARE DEFECTS
csandit
 
Srushti_M.E_PPT.ppt
Srushti_M.E_PPT.pptSrushti_M.E_PPT.ppt
Srushti_M.E_PPT.ppt
khalid aberbach
 
IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...
IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...
IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...
IRJET Journal
 

Similar to Ajila (1) (20)

A scalable collaborative filtering framework based on co-clustering
A scalable collaborative filtering framework based on co-clusteringA scalable collaborative filtering framework based on co-clustering
A scalable collaborative filtering framework based on co-clustering
 
Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013
 
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
 
An Adaptive Load Balancing Middleware for Distributed Simulation
An Adaptive Load Balancing Middleware for Distributed SimulationAn Adaptive Load Balancing Middleware for Distributed Simulation
An Adaptive Load Balancing Middleware for Distributed Simulation
 
A Hierarchical Feature Set optimization for effective code change based Defec...
A Hierarchical Feature Set optimization for effective code change based Defec...A Hierarchical Feature Set optimization for effective code change based Defec...
A Hierarchical Feature Set optimization for effective code change based Defec...
 
Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)
Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)
Machine learning in Dynamic Adaptive Streaming over HTTP (DASH)
 
Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)
Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)
Dependable Systems - Structure-Based Dependabiilty Modeling (6/16)
 
IRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware PerformanceIRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware Performance
 
IRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor DriveIRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
 
Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...
Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...
Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Sched...
 
Thesis Presentation on Energy Efficiency Improvement in Data Centers
Thesis Presentation on Energy Efficiency Improvement in Data CentersThesis Presentation on Energy Efficiency Improvement in Data Centers
Thesis Presentation on Energy Efficiency Improvement in Data Centers
 
Svm on cloud (presntation)
Svm on cloud  (presntation)Svm on cloud  (presntation)
Svm on cloud (presntation)
 
Svm Classifier Algorithm for Data Stream Mining Using Hive and R
Svm Classifier Algorithm for Data Stream Mining Using Hive and RSvm Classifier Algorithm for Data Stream Mining Using Hive and R
Svm Classifier Algorithm for Data Stream Mining Using Hive and R
 
Maestro_Abstract
Maestro_AbstractMaestro_Abstract
Maestro_Abstract
 
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
 
Configuration Navigation Analysis Model for Regression Test Case Prioritization
Configuration Navigation Analysis Model for Regression Test Case PrioritizationConfiguration Navigation Analysis Model for Regression Test Case Prioritization
Configuration Navigation Analysis Model for Regression Test Case Prioritization
 
M3AT: Monitoring Agents Assignment Model for the Data-Intensive Applications
M3AT: Monitoring Agents Assignment Model for the Data-Intensive ApplicationsM3AT: Monitoring Agents Assignment Model for the Data-Intensive Applications
M3AT: Monitoring Agents Assignment Model for the Data-Intensive Applications
 
ESTIMATING HANDLING TIME OF SOFTWARE DEFECTS
ESTIMATING HANDLING TIME OF SOFTWARE DEFECTSESTIMATING HANDLING TIME OF SOFTWARE DEFECTS
ESTIMATING HANDLING TIME OF SOFTWARE DEFECTS
 
Srushti_M.E_PPT.ppt
Srushti_M.E_PPT.pptSrushti_M.E_PPT.ppt
Srushti_M.E_PPT.ppt
 
IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...
IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...
IRJET- Enhance Dynamic Heterogeneous Shortest Job first (DHSJF): A Task Schedu...
 

Recently uploaded

Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
Jason Packer
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
Zilliz
 
The Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptxThe Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptx
operationspcvita
 
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
Fwdays
 
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing InstancesEnergy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Alpen-Adria-Universität
 
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development ProvidersYour One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
akankshawande
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
Jakub Marek
 
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansBiomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Neo4j
 
HCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAUHCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAU
panagenda
 
Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...
Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...
Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...
Pitangent Analytics & Technology Solutions Pvt. Ltd
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
DianaGray10
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
Edge AI and Vision Alliance
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
Chart Kalyan
 
GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)
Javier Junquera
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
Miro Wengner
 
Nordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptxNordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptx
MichaelKnudsen27
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
Tatiana Kojar
 
Harnessing the Power of NLP and Knowledge Graphs for Opioid Research
Harnessing the Power of NLP and Knowledge Graphs for Opioid ResearchHarnessing the Power of NLP and Knowledge Graphs for Opioid Research
Harnessing the Power of NLP and Knowledge Graphs for Opioid Research
Neo4j
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
AstuteBusiness
 
What is an RPA CoE? Session 1 – CoE Vision
What is an RPA CoE?  Session 1 – CoE VisionWhat is an RPA CoE?  Session 1 – CoE Vision
What is an RPA CoE? Session 1 – CoE Vision
DianaGray10
 

Recently uploaded (20)

Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
 
The Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptxThe Microsoft 365 Migration Tutorial For Beginner.pptx
The Microsoft 365 Migration Tutorial For Beginner.pptx
 
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
 
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing InstancesEnergy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
 
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development ProvidersYour One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
 
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansBiomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
 
HCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAUHCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAU
 
Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...
Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...
Crafting Excellence: A Comprehensive Guide to iOS Mobile App Development Serv...
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
 
GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
 
Nordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptxNordic Marketo Engage User Group_June 13_ 2024.pptx
Nordic Marketo Engage User Group_June 13_ 2024.pptx
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
 
Harnessing the Power of NLP and Knowledge Graphs for Opioid Research
Harnessing the Power of NLP and Knowledge Graphs for Opioid ResearchHarnessing the Power of NLP and Knowledge Graphs for Opioid Research
Harnessing the Power of NLP and Knowledge Graphs for Opioid Research
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
 
What is an RPA CoE? Session 1 – CoE Vision
What is an RPA CoE?  Session 1 – CoE VisionWhat is an RPA CoE?  Session 1 – CoE Vision
What is an RPA CoE? Session 1 – CoE Vision
 

Ajila (1)

  • 1. Proactive Prediction Models for Web Application Resource Provisioning in the Cloud _______________________________ Samuel A. Ajila & Bankole A. Akindele
  • 2. Presentation Outline  Introduction to Problem Area  Motivation  Goals and Scope  Contributions  Related work  Machine learning algorithms  Implementation setup  Evaluation metrics  Selected results  Conclusion 2
  • 3. Introduction  Cloud computing refers to both the applications delivered as services over the Internet and the hardware and systems software in the data centers that provide those services (SaaS, PaaS and IaaS)  Challenges include: Data security threats, performance unpredictability and prompt (quick) resource scaling  Accurate Virtual machine (VM) resource provisioning: • under and over-provisioning  Current techniques like Control theory, Constraint programming and Machine learning 3
  • 4. Motivation  VM instantiation (scaling) takes time and ranges from 5 to 12 minutes  Challenges that can result from the instantiation duration: • Possibility of Service Level Agreement (SLA) violations - Cloud providers • Poor customer’s Quality of Experience (QoE) – Cloud client • Reputation loss – Cloud client/providers  Presently, monitoring metrics made available to clients are limited to CPU, Memory and Network utilization  The motivation is to “predict resource usage so that cloud providers can make adequate provisioning ahead of time”  To extend the monitoring metrics by including Response time and Throughput 4
  • 5. Goals and Scope  Design and develop a cloud client prediction model for cloud resource provisioning in a Multitier web application environment  The model would be capable of forecasting future resource usage to enable timely VM provisioning  To achieve this goal, SVM, NN and LR learning techniques are analysed using the Java implementation of TPC-W (workload)  The scope of this work is limited to IaaS  Prediction model is built around the web server tier. It is possible to extend to other tiers 5
  • 6. Contributions  Design and development of a cloud client prediction model that uses historical data to forecast future resource usage  The evaluation of the resource usage prediction capability of SVM, NN and LR using three benchmark workloads from TPC-W  The extension of the prediction model to include Throughput and Response time, thus providing wider and better scaling decision options for cloud clients  The comparison of the prediction capability of SVM, NN and LR models under random and steady traffic patterns 6
  • 7. Related works Table 1: Auto-scaling techniques 7 Technique Class Comments Threshold/Rule based Reactive Reacts to system changes but do not anticipate them Control theory Reactive/Proactive Except where used with a proactive approach it suffers reactive issues Queuing theory Reactive As complexity of system grows, the analytic formulas become difficult Reinforcement learning Reactive/Proactive Good but converging at optimal policy can be unfeasibly long (state –action pairs) Time series analysis Proactive Good and promising
  • 8. Machine learning algorithms Linear Regression  The linear regression model has the form: 𝑓 𝑋 = 𝛽0 + 𝑗=1 𝑝 𝑋𝑗 𝛽𝑗 (1)  𝑋𝑗 is the input data  𝛽0 and 𝛽𝑗’s are the unknown parameters estimated from the training data. Estimation is done using the least squares method.  The coefficients 𝛽 are picked to minimize the residual (difference between the actual and predicted value) sum of squares : (𝑦𝑖 − 𝑓 𝑋 )2 𝑦𝑖 is the actual value 8
  • 9. Machine learning algorithms (cont’d) Neural Network  A two-stage regression or classification model represented by a network diagram  Derived features 𝑍 𝑚 are created from linear combinations of the input after which the target 𝑌𝑘 is modeled as a linear combination of 𝑍 𝑚  Like Linear regression, unknown parameters called weights are sought to make the model fit the training data well 9 Figure 1 : Single hidden layer, feed forward neural network
  • 10. Machine learning algorithms (cont’d) Support Vector Regression  The goal of SVR is to find a function 𝑓 𝑋 that has at most 𝜀 (the precision by which the function is to be approximated) deviation from the actual obtained target for all training data  Mathematically, 𝑓 𝑋 = 𝑤. ∅ 𝑥 + 𝑏  Input data 𝑥 is mapped to higher dimensional feature space via the kernel function, then a linear regression is performed  Goal is to find optimal weights 𝑤 and threshold 𝑏 10
  • 12. Implementation setup (cont’d) Time (minutes) 1-7 56-63 154-161 350-357 490-497 504-511 Shopping mix users 84 168 16 180 248 160 Browsing mix users 52 112 36 320 192 160 Ordering mix users 52 108 28 224 268 160 Total user Requests 188 388 80 724 708 480 12  Total length of experiment was about 10 hours  Selected experimental workload mix
  • 13. Evaluation metrics 13  Evaluation is done on the 60% training and 40% held out test dataset  The held out dataset is used to forecast to a maximum interval of 12 minutes (VM instantiation time reported by other authors) Performance metrics and their calculations Metric Calculation MAPE (Mean Absolute Percentage Error) 1 𝑛 | 𝑎𝑖−𝑝𝑖| 𝑎𝑖 𝑛 𝑖=1 where 𝑎𝑖 and 𝑝𝑖 are the actual and predicted values respectively RMSE (Root Mean Square Error) √ ( 𝑎𝑖−𝑝𝑖) 2𝑛 𝑖=1 𝑛 MAE (Mean Absolute Error) 1 𝑛 |𝑝𝑖 − 𝑎𝑖|𝑛 𝑖=1 PRED 25 No. of observations with relative error ≤ 25% / No. of observation
  • 14. CPU Training & Testing Performance 14 CPUUtilizationTrainingandTestingPerformanceMetricforLR,NN,andSVR LR(LinearRegression)Model MAPE RMSE MAE PRED(25) LRTraining 113.31 14.70 11.11 0.51 LRTesting 36.19 22.13 15.98 0.36 NN(NeuralNetwork)Model MAPE RMSE MAE PRED(25) NNTraining 105.63 14.08 9.48 0.59 NNTesting 50.46 31.08 19.82 0.34 SVR(SupportVectorRegression)Model MAPE RMSE MAE PRED(25) SVRTraining 107.80 15.48 10.09 0.64 SVRTesting 22.84 11.84 8.74 0.64
  • 15. Selected results (cont’d) 15 CPU utilization test performance metric Model MAPE RMSE MAE PRED(25) LR 36.19 22.13 15.98 0.36 NN 50.46 31.08 19.82 0.34 SVR 22.84 11.84 8.74 0.64 Figure 5 CPU Utilization Actual and Predicted test model results
  • 16. Selected results (cont’d) 16 Throughput test performance metric Model MAPE RMSE MAE PRED(25) LR 24.62 3.72 2.87 0.63 NN 38.90 6.12 4.46 0.47 SVR 22.07 3.22 2.41 0.67 Figure 6 Throughput Actual and Predicted test model results
  • 17. Selected results (cont’d) 17 Response time test performance metric Model MAPE RMSE MAE PRED(25) LR 12.35 1.39 1.11 0.91 NN 17.84 2.02 1.64 0.75 SVR 9.92 1.21 0.87 0.93 Figure 7 Response time Actual and Predicted test model results
  • 18. CPU - Comparison of Prediction Models 18
  • 19. Throughput – Comparison of Prediction Models 19
  • 20. Sensitivity Measurement Using Little’s Law 20 Data Consistency Measurement Using Little’sLaw Time (minute) 1-7 56-63 154- 161 350- 357 490- 497 498- 503 504- 511 Average total user requests 27 (188/7) 55 (388/7) 11 (80/7) 103 (724/7) 101 (708/7) 95 (664/7) 69 (480/7) Average Throughput (Requests/second) 5.29 10.52 2.47 15.92 10.00 12.23 13.27 Average time spent (seconds) 5.10 5.23 4.45 6.47 10.01 7.77 5.2 Measured time spent 4.66 4.92 9.92 7.53 9.06 9.01 9.45 Time variance (%) 9.44 6.31 55.14 14.08 10.49 13.76 44.97
  • 21. Conclusion  SVR displayed superior prediction accuracy over both LR and NN in a typically nonlinear, not defined a-priori workload by:  In the CPU utilization prediction model, SVR outperformed LR and NN by 58% and 120% respectively  For the Throughput prediction model, SVR again outperformed LR and NN by 12% and 76% respectively; and finally,  The Response time prediction model saw SVR outperforming LR and NN by 26% and 80% respectively.  Based on this experimental results SVR may be accepted as the best prediction model in a nonlinear, not defined a-priori system 21
  • 22. Future works  SVR and other machine learning algorithms are good for forecasting, however • training and retraining is a challenge • Parameter selection is still empirical  Combination of SVR and other predicting techniques may mitigate this challenge  Other future direction include • Inclusion of database tier for a more robust scaling decision • Resource prediction on other non-web application workload 22