modeling and exploiting qos prediction in cloud and service computing yilei zhang thesis committee...
TRANSCRIPT
![Page 1: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/1.jpg)
Modeling and Exploiting QoS Prediction
in Cloud and Service Computing
Yilei Zhang
Thesis CommitteeChair: Prof. Pak Ching LEE
Supervisor: Prof. Michael R. LYUMember: Prof. Fung Yu YOUNGExternal Examiner: Prof. Qing LI
Sep. 12, 2013
![Page 2: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/2.jpg)
Outlines
• Introduction• Part1: QoS Prediction Approaches
– Neighborhood-Based Approach– Time-aware Model-Based Approach– Online Approach
• Part2: QoS-Aware Web Service Searching• Part3: QoS-Aware Byzantine Fault Tolerance• Conclusion
2
![Page 3: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/3.jpg)
Cloud Computing
• Cloud component– Software, server, database, etc.
• On-demand
3
![Page 4: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/4.jpg)
Cloud Applications
• Software-as-a-Service (SaaS) – Large-scale, complicated, time
sensitive, high-quality
• Case 1: New York Times– Convert scanned articles to PDF– 15 million files, 4TB data– EC2 & S3, 100 computers 24 hours
• Case 2: Nasdaq– Stock and fund information– Millions of files, per 10 minutes– S3
4
![Page 5: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/5.jpg)
Web Services
• Web APIs– Accessed over a network,– Executed on remote systems– Loosely-coupled– Compositional nature
5
![Page 6: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/6.jpg)
Service-Oriented-Architecture
• Service-Oriented-Architecture (SOA) – Distributed Web services– 30,000 services, 200,000 documents
(seekda.com)Hotel Web Services
Airline Web Services
Credit Card Web
Services
Travel Agency Web Site 1
Travel Agency Web Site 2
6
![Page 7: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/7.jpg)
Performance of Services
• Service– Web service– Cloud component
• User observed performance.– Remote network
access– Location – Invocation time
7
![Page 8: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/8.jpg)
Quality-of-Service
• Quality-of-Service (QoS): non-functional performance– User/Time-independent QoS properties
• price, popularity
– User/Time-dependent QoS properties• failure probability, response time, throughput
• High quality applications depends on high quality services– Service selection, service searching, fault
tolerance, service composition, etc.
8
![Page 9: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/9.jpg)
Challenge 1: How to Obtain QoS?
• Conducting real-world evaluations?• Drawbacks
– Expensive (charge for real invocations) – Time-consuming (thousands of services)– Personalized evaluation (users’ perspective)– Expertise (extra cost and effort)
9
![Page 10: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/10.jpg)
Challenge 1: How to Obtain QoS?
• Solution: QoS Prediction (Part 1)– Collect users’ usage experiences – Analyze historical QoS data– Predict QoS values
• Advantages– Economical (no additional invocation)– Precise (personalized QoS)– Effective (no extra expertise)– Efficient (provided as a service)
10
![Page 11: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/11.jpg)
Challenge 2: How to Search Appropriate Services?
• Problems– Thousands of services– Different QoS Performance
• Solution: QoS-aware searching (Part 2)
11
![Page 12: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/12.jpg)
Challenge 3: How to Build Reliable Service-Oriented
Systems
• Problems– Services may contain various faults– QoS of remote services may not be stable, e.g.,
unavailability problem
• Solution: QoS-aware fault tolerance (Part 3)
12
![Page 13: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/13.jpg)
Thesis Structure
13
Part 1: QoS Prediction
Part 2: QoS-Aware Searching
Part 3: QoS-Aware Fault Tolerance
Chapter 7Chapter 6
Chapter 5Chapter 4Chapter 3
![Page 14: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/14.jpg)
Approach 1Neighborhood-based QoS
Prediction
Part 1: QoS Prediction
Part 2: QoS-Aware Searching
Part 3: QoS-Aware Fault Tolerance
Chapter 7Chapter 6
Chapter 5Chapter 4Chapter 3
![Page 15: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/15.jpg)
Toy Example
• User-component matrix: m × n, each entry is a QoS value.– Sparse– Prediction accuracy is greatly influenced by
similarity computation.
15
![Page 16: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/16.jpg)
Latent Features Learning
16
Latent-user matrix V
u1 u2 u3 u4
Latent-component matrix H
c1 c2 c3 c4 c5 c6
W = V X HT
![Page 17: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/17.jpg)
Similarity Computation
• Pearson Correlation Coefficient (PCC) • Similarity between users:
• Similarity between components:
Latent-component matrix H
Latent-user matrix V
u1 u2 u3 u4
c1 c2 c3 c4 c5 c6
17
![Page 18: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/18.jpg)
Neighbors Selection
• For every entry wi,j in the matrix, a set of similar users towards user ui can be found by:
• A set of similar items towards component cj can be found by:
18
![Page 19: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/19.jpg)
Missing Value Prediction
• Similar User-based:
• Similar Component-based:
• Hybrid:
19
![Page 20: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/20.jpg)
Experiments
• QoS Dataset
• Metrices– Mean Absolute Error (MAE)– Root Mean Squared Error (RMSE)
– : the real QoS value.– : the predicted QoS value– N: the number of predicted values.
20
![Page 21: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/21.jpg)
Performance Comparisons
• IPCC – similar item
• UPCC – similar user
• UIPCC– similar item + similar user
• NMF– matrix factorization
• CloudPred– matrix factorization + similar item + similar user
21
![Page 22: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/22.jpg)
Experimental Results
22
![Page 23: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/23.jpg)
Impact of Matrix Density
23
![Page 24: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/24.jpg)
Impact of Top-K
24
![Page 25: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/25.jpg)
Impact of Dimensionality
25
![Page 26: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/26.jpg)
Part 1: QoS Prediction
Part 2: QoS-Aware Searching
Part 3: QoS-Aware Fault Tolerance
Chapter 7Chapter 6
Chapter 5Chapter 4Chapter 3
Approach 2Time-Aware Model-Based QoS Prediction
![Page 27: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/27.jpg)
Time-Aware QoS Performance
• Time-aware personalized QoS prediction is essential for:– Automatic selection– Dynamic composition
27
![Page 28: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/28.jpg)
Case Study
• Periodic feature + average performance
28
![Page 29: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/29.jpg)
Tensor FactorizationTime
Service
latent-service matrixlatent-user matrix
u1 u2 u3 u4 u5
latent-time matrix
s1 s2 s3 s4 s5 t1 t2 t3 t4 t5
29
User
![Page 30: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/30.jpg)
Latent Features Learning
• Objective function
• Local optimal solution is found by incremental gradient descent
The error between estimated tensor and the original tensor
Regularization terms which constrain the norms of U, S and T, to avoid overfitting problem
Regularization term which prevents the predicted QoS values from varying a lot against the average QoS value
30
![Page 31: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/31.jpg)
Missing Value Prediction
• Given feature spaces U, S and T
• Example:
latent-service matrixlatent-user matrix
u1 u2 u3 u4 u5
latent-time matrix
s1 s2 s3 s4 s5 t1 t2 t3 t4 t5
341Y
31
![Page 32: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/32.jpg)
Experiments
• Time-Aware Web Service QoS Dataset
32
![Page 33: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/33.jpg)
Performance Comparisons
• Matrix Factorization extended methods– MF1: a set of user-service matrix slices in terms
of time– MF2: compresses the user-service-time tensor
into an user-service matrix
• Tensor Factorization methods– TF :Tensor factorization-based prediction
method.– WSPred :Tensor factorization-based prediction
method with average QoS value constraints.
33
![Page 34: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/34.jpg)
Experimental Results
• A smaller MAE or RMSE value means a better performance
9~25% 5~15% 3~12% 16~22% 3~13% 1~12%
Performance improvement of WSPred
34
![Page 35: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/35.jpg)
Impact of Tensor Density
35
![Page 36: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/36.jpg)
Impact of Dimensionality
36
![Page 37: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/37.jpg)
Approach 3Online QoS Prediction
Part 1: QoS Prediction
Part 2: QoS-Aware Searching
Part 3: QoS-Aware Fault Tolerance
Chapter 7Chapter 6
Chapter 5Chapter 4Chapter 3
![Page 38: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/38.jpg)
System Architecture
38Online QoS Prediction Procedures
![Page 39: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/39.jpg)
Key Idea
39
t1 … tc-1
tc
![Page 40: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/40.jpg)
Step 1: Time-Aware Latent Feature Learning
• Objective function:
The error between estimated matrix and the original matrix
Regularization terms which constrain the norms of p(t) and q(t), to avoid overfitting problem
40
![Page 41: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/41.jpg)
Step 1: Time-Aware Latent Feature Learning
• Iterative Process :– gradient descent
41
![Page 42: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/42.jpg)
Step 2 & Step 3 (Offline Phase)
u1 u2 u3 u4 u5
c1 c2 c3 c4 c5
42
![Page 43: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/43.jpg)
Step 2 & Step 3 (Online Phase)
43
![Page 44: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/44.jpg)
System Level Performance
44
Basic Compositional Structures
Calculation of Aggregated Response Time
![Page 45: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/45.jpg)
System Level Performance Prediction
45
![Page 46: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/46.jpg)
Comparison with Other Methods
• UPCC (time-insensitive) - Mean• IPCC (time-insensitive) - Mean• MF (time-insensitive) - Mean• TF (time-sensitive) - Periodic • WSPred (time-sensitive) – Periodic + Mean• OPred (time-sensitive) - Periodic + Mean +
timely trend
46
![Page 47: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/47.jpg)
Experimental Results
47
![Page 48: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/48.jpg)
Impact of Density
48
![Page 49: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/49.jpg)
Impact of Dimensionality
49
![Page 50: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/50.jpg)
Impact of a and w
50
![Page 51: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/51.jpg)
Average Computational Time
51
![Page 52: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/52.jpg)
System Level Performance Case Study
52
![Page 53: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/53.jpg)
System Level Performance Case Study
53
System Performance Improvement of Dynamic Service Composition
![Page 54: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/54.jpg)
Part 1: QoS Prediction
Part 2: QoS-Aware Searching
Part 3: QoS-Aware Fault Tolerance
Chapter 7Chapter 6
Chapter 5Chapter 4Chapter 3
![Page 55: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/55.jpg)
Problems
• How to find a service?– Many Web services
• How to find the best one?– Different QoS Performance
Functionality
QoS
55
![Page 56: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/56.jpg)
WSExpress
• Functional attributes & non-functional features.– Non-functional evaluation
• Obtains QoS criteria values• QoS utility computation
– Functional evaluation• WSDL preprocessing• Similarity computation
56
![Page 57: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/57.jpg)
Combination
57
![Page 58: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/58.jpg)
Performance Comparisons
A larger NDCG value means a better performance58
![Page 59: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/59.jpg)
Contributions
• Functionality and non-functionality• A large-scale distributed experimental
evaluation– 3738 Web services– 69 countries
• real-world WSDL dataset and QoS dataset– 30+ institutes
59
![Page 60: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/60.jpg)
Part 1: QoS Prediction
Part 2: QoS-Aware Searching
Part 3: QoS-Aware Fault Tolerance
Chapter 7Chapter 6
Chapter 5Chapter 4Chapter 3
![Page 61: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/61.jpg)
How to Build Reliable Service-Oriented Systems
• Problems– Services may contain various faults– QoS of remote services may not be stable, e.g.,
unavailability problem
• Solution: QoS-aware fault tolerance
61
![Page 62: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/62.jpg)
System Architecture
62
BFT Group
![Page 63: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/63.jpg)
Work Procedures of BFTCloud
63
![Page 64: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/64.jpg)
Experimental Results
64Request/Response Size: 0/4 KB
Request/Response Size: 0/0 KB Request/Response Size: 4/0 KB
![Page 65: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/65.jpg)
Contributions
• A Byzantine fault tolerance framework– QoS-aware
• A prototype system• Large-scale real-world experiments
65
![Page 66: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/66.jpg)
Thesis Structure
66
Part 1: QoS Prediction
Part 2: QoS-Aware Searching
Part 3: QoS-Aware Fault Tolerance
Chapter 7Chapter 6
Chapter 5Chapter 4Chapter 3
![Page 67: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/67.jpg)
Conclusion
• Part 1: QoS prediction– Three prediction approaches– Several following up works on this topic using
the released datasets
• Prat 2: QoS-aware service searching– Searching qualities are significantly improved
• Part 3: QoS-aware fault tolerance– Byzantine fault tolerance– dynamic QoS information
67
![Page 68: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/68.jpg)
Publications
Paper published• Yilei Zhang, Zibin Zheng and Michael R. Lyu. "Real-Time Performance Prediction for
Cloud Components", in Proceedings of the 5th International Workshop on Real-Time Service-Oriented Architecture and Applications (RTSOAA 2012), Shenzhen, China, Apr. 11-Apr. 13, 2012, pp.106-111.
• Yilei Zhang, Zibin Zheng and Michael R. Lyu. "WSPred: A Time-Aware Personalized QoS Prediction Framework for Web Services", in Proceedings of the 22th IEEE Symposium on Software Reliability Engineering (ISSRE 2011), Hiroshima, Japan, Nov. 29-Dec. 2, 2011, pp.210-219.
• Yilei Zhang, Zibin Zheng and Michael R. Lyu. "Exploring Latent Features for Memory-Based QoS Prediction in Cloud Computing", in Proceedings of the 30th IEEE Symposium on Reliable Distributed Systems (SRDS 2011), Madrid, Spain, Oct. 4-7, 2011, pp.1-10.
• Yilei Zhang, Zibin Zheng and Michael R. Lyu. "BFTCloud: A Byzantine Fault Tolerance Frame-work for Voluntary-Resource Cloud Computing", in Proceedings of the 4th IEEE International Conference on Cloud Computing (CLOUD 2011), Washington DC, USA, July 4-9, 2011, pp.444-451.
• Zibin Zheng, Yilei Zhang, and Michael R. Lyu, “Investigating QoS of Real-World Web Services”, IEEE Transactions on Service Computing.
68
![Page 69: Modeling and Exploiting QoS Prediction in Cloud and Service Computing Yilei Zhang Thesis Committee Chair: Prof. Pak Ching LEE Supervisor: Prof. Michael](https://reader035.vdocuments.us/reader035/viewer/2022062716/56649e165503460f94b00f50/html5/thumbnails/69.jpg)
Publications
• Yilei Zhang, Zibin Zheng and Michael R. Lyu. "WSExpress: A QoS-Aware Search Engine for Web Services", in Proceedings of the 8th IEEE International Conference on Web Services (ICWS 2010), Miami, Florida, USA, July 5-10, 2010, pp. 91-98.
• Zibin Zheng, Xinmiao Wu, Yilei Zhang, Michael R. Lyu and Jianmin Wang. "QoS Ranking Prediction for Cloud Services", IEEE Transactions on Parallel and Distributed Systems.
• Zibin Zheng, Yilei Zhang and Michael R. Lyu. "Distributed QoS Evaluation for Real-World Web Services”, in Proceedings of the 8th IEEE International Conference on Web Services (ICWS 2010), Miami, Florida, USA, July 5-10, 2010, pp. 83-90.
• Zibin Zheng, Yilei Zhang and Michael R. Lyu. "CloudRank: A QoS-Driven Component Ranking Framework for Cloud Computing", in proceedings of the 28th IEEE International Symposium on Reliable Distributed Systems (SRDS 2010), New Delhi, India, Oct.31-Nov.3, 2010.
Paper under review/preparation• Yilei Zhang, Zibin Zheng and Michael R. Lyu. "An Online Performance Prediction
Framework for Service-Oriented Systems", submitted to the IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans.
• Yilei Zhang, Zibin Zheng and Michael R. Lyu. "QoS-Aware Web Service Searching ", prepared to submit to IEEE Transactions on Service Computing.
• Yilei Zhang, Zibin Zheng and Michael R. Lyu. “QoS Prediction via Latent Feature Learning in Cloud Computing", prepared to submit to IEEE Transactions on Cloud Computing.
69