Distributed Architectures with Genetic Algorithms

Transcript

1. Finding the Best Configuration in Distributed Architectures through Genetic Algorithms

2. Passionate by Chaos Engineering YURY NIÑO ROA Cloud Infrastructure Engineer

   Site Reliability Engineer Advocate www.sitereliabilityengineering.co www.ingenieriadelcaos.com www.yurynino.com

3. 03 The problem! The Solution! 02 Genetics Algorithms 01 AGENDA

   The problem! Distributed Architectures Circuit Breaker

4. DISTRIBUTED ARCHITECTURES

5. DISTRIBUTED ARCHITECTURES 1990s 2000s 2010s PRE-SOA Monolithics Tight Coupling Each

   change has unanticipated effects. Traditional SOA Looser Coupling Components must be coordinated. Microservices Decoupled Components take advantage of CI/CD www.yurynino.com www.sitereliabilityengineering.co

6. DISTRIBUTED ARCHITECTURES That is the promise But the reality is

   … Netflix Twitter www.yurynino.com www.sitereliabilityengineering.co

7. The infrastructure required by a software system can be as

   complex as the architecture itself. Every production failure is unique. www.yurynino.com www.sitereliabilityengineering.co

8. Butterfly Spiderman THE PROBLEM www.yurynino.com www.sitereliabilityengineering.co

9. Slow Responses Cascading Failures MORE PROBLEMS www.yurynino.com www.sitereliabilityengineering.co

10. THE SOLUTION www.yurynino.com www.sitereliabilityengineering.co

11. CIRCUIT BREAKER HAS A CHALLENGE

12. CIRCUIT BREAKER Taken from Release it! www.yurynino.com www.sitereliabilityengineering.co

13. THE CHALLENGE Taken from Release it! www.yurynino.com www.sitereliabilityengineering.co

14. WHAT IS A PROPER CONFIGURATION?

15. GENETIC ALGORITHMS

16. Inspired by the Evolutionary Computation By imitating the process of

    natural selection and reproduction, genetic algorithms can produce high-quality solutions for various problems involving search, optimization, and learning. www.yurynino.com www.sitereliabilityengineering.co

17. Genetic Algorithms seek to find the optimal solution for a

    given problem. • Maintain a population of candidate solutions, called individuals, for that given problem. • These candidate solutions are iteratively evaluated and used to create a new generation of solutions. • Those who are better at solving this problem have a greater chance of being selected and passing to the next generation!

18. Initial Population Evaluate Selection Crossover Mutation Termination Criterion? Solution Set

    THE METHODOLOGY

19. Initial Population Selection Crossover THE METHODOLOGY

20. Initial Population Evalua te Selection Mutation THE METHODOLOGY Evaluate

21. Initial Population Evaluate Selection Crossover Mutation Termination Criterion? Solution Set

    THE METHODOLOGY

22. Mutation THE METHODOLOGY

23. Initial Population Evaluate Selection Crossover Mutation Termination Criterion? Solution Set

    THE METHODOLOGY

24. “It is not the strongest of the species that survives,

    nor the most intelligent, but the one who is most adaptable to change.” Charles Darwin

25. Thanks! Yury Niño Roa www.yurynino.com yurynino