Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems
description
Transcript of Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems
![Page 1: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/1.jpg)
Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems
Authors:Andres J. Ramirez, David B. Knoester, Betty H.C. Cheng, Philip K. McKinley
Michigan State University
Presented By:Shivashis Saha
University of Nebraska-Lincoln
![Page 2: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/2.jpg)
2
Outline• Introduction• Background – Remote Mirroring– Genetic Algorithm
• Proposed Approach– Plato Design– Fitness Function
• Case Study• Conclusion
![Page 3: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/3.jpg)
3
Introduction
• Autonomic Computing System– What is it?• Self-configurable• Anticipated execution vs Dynamic reconfiguration
• Three key components:1. Monitoring2. Decision making3. Reconfiguration
![Page 4: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/4.jpg)
4
Introduction
• Reconfiguration– Rule based decision making– Utility based decision making
Self adapt to scenarios considered at design time– Evolutionary computations
Limited to specific set of reconfiguration strategies
![Page 5: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/5.jpg)
5
Contributions• Plato – GA based decision making process• Reconfiguration plans for changing requirements and
environmental conditions No need to plan in advance
• Dynamic reconfiguration of an overlay network • Distributing data to a collection of remote data
mirrors• Design Objectives:
1. Minimize cost2. Maximize data reliability3. Maximize network performance
![Page 6: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/6.jpg)
6
Remote Mirroring
• Copies of important data are stored at one or more secondary locations– Tradeoff between better performance with lower
cost against greater potential for data loss• Design choices– Type of links• Throughput, latency, loss rate
– Remote mirroring protocols• Synchronous vs Asynchronous
![Page 7: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/7.jpg)
7
Genetic Algorithm
• Chromosomes– <AB, BC, CD, AD, AC, BD>
![Page 8: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/8.jpg)
8
Genetic Algorithm• Crossover– One-point vs Two-point
![Page 9: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/9.jpg)
9
Genetic Algorithm• Mutation
![Page 10: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/10.jpg)
10
Genetic Algorithm
• Summarize the entire approach:1. Generate initial population2. Use crossover and mutation to generate new
generation3. Evaluate the offspring4. Survival of the fittest5. Go to step 2; Terminate when desired value
achieved or the algorithm converged
![Page 11: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/11.jpg)
11
Proposed Approach
• Input: A network of remote data mirrors• Output: Construct an overlay network– Data can be distributed to all the nodes– Network must remain connected– Never exceed monetary budget– Minimize bandwidth for diffusing data
![Page 12: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/12.jpg)
12
Plato Design• Chromosome encodes a complete network– Link
• active or inactive• Associated with seven propagation method
• There are 7n(n-1)/2*2n(n-1)/2 possible link configurations
![Page 13: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/13.jpg)
13
Fitness Function
Network’s fitness in terms of cost
Network’s fitness in terms of performance
Network’s fitness in terms of reliability
![Page 14: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/14.jpg)
14
Fitness Function
Operational costs incurred by all active links
Maximum amount of money allocated for the network
![Page 15: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/15.jpg)
15
Fitness Function
Average latency over all active links
Largest measured latency in the underlying network
Total bandwidth in the overlay network
Total effective bandwidth in the overlay network after data has been coalesced
Limit on the best value achieved in terms of bandwidth reduction
![Page 16: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/16.jpg)
16
Fitness Function
Number of active links used Maximum number of possible links
Total amount of data that could be lost as a result of the propagation method
Amount of data that could be lost by selecting the propagation method with the largest time window for write coalescing
![Page 17: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/17.jpg)
17
Fitness Function
• Each link stores the values such as throughput, latency, loss rate, etc
• Rescale coefficients when requirements change– The change of coefficients are mentioned in
response to high-level monitoring events– Do not explicitly specify the reconfiguration plan
![Page 18: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/18.jpg)
18
Case Study
![Page 19: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/19.jpg)
19
Case Study
GA converges
![Page 20: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/20.jpg)
20
Case Study
1. Maximum fitness around 88, not 100!2. Activate all 300 links for maximum reliability3. Synchronous propagation is dominant
![Page 21: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/21.jpg)
21
Case Study
1. Network has 32 links, majority use asynchronous propagation2. Overall, provides a combination of performance and reliability
while keeping the cost low
![Page 22: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/22.jpg)
22
Case Study
![Page 23: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/23.jpg)
23
Case StudyAt first more fit network has fewest active links. But, at the end 8 additional links were added.
![Page 24: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/24.jpg)
24
Case Study
![Page 25: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/25.jpg)
25
Case Study
![Page 26: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/26.jpg)
26
Case Study
Link failure
![Page 27: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/27.jpg)
27
Case Study
Link failureMinimize cost
Improved robustness
![Page 28: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/28.jpg)
28
Case Study
Data loss: measured on a log scale; byproduct of the propagation methods
Link failure
![Page 29: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/29.jpg)
29
Conclusion• Plato integrates GA into decision making
process of adaptive and autonomic systems– Supports dynamic reconfiguration– Does not explicitly encode prescriptive
reconfiguration strategies to address scenarios which may arise in future
– It uses user defined fitness to evolve reconfiguration plans in response to environmental changes
![Page 30: Applying Genetic Algorithms to Decision Making in Autonomic Computing Systems](https://reader036.fdocuments.net/reader036/viewer/2022070423/568167f5550346895ddd6ef3/html5/thumbnails/30.jpg)
30
Thanks!