Fireflies, Ants and Bees! Mother Nature builds effective, fault tolerant, distributed systems!

A tour though biologically inspired metahuristic algorithms and how specifically ACO+TSP could apply to data replication.

Transcript

1. Fireflies, Ants and Bees! Mother Nature builds effective, fault tolerant, distributed systems!
2. Distributed Realtime requirements. not features... Many networked programs that make up a larger application. Minimum cost paths.
3. distributed realtime ...they seem like opposing forces.
4. Mother Nature is one clever engineer
5. Fireflies!
7. Firefly Swarm Optimization Algorithm Application Distributed Fault Detection
8. Fault identification with binary adaptive fireflies in parallel and distributed systems R. Falcon A Hybrid Firefly-inspired Approach for Optimal Semantic Web Service Composition Cristina Pop, et al.
9. Bees!
10. Algorithm Application Efficient routing in ad-hoc mobile networks Artificial bee colony algorithm
11. Formal Modeling of BeeAdHoc: a Bio-inspired Mobile Ad Hoc Network Routing Protocol Muhammad Saleem, et al.
12. Ants!
13. Algorithm Application Information Foraging, Traveling Salesman Problem Ant Colony Optimization
14. Ant Colony Optimization Marco Dorigo and Thomas Stützle
15. Double-Bridge Experiment Minimum Cost Paths
16. Traveling Salesman Problem {c1, c2, ... cN} {ci, cj } d(ci, cj ) (pair) (cities) (distance)
17. effective node selection for data replication
18. Random Node Selection Naive, Discovery Phase
19. Intentional Node Selection ACO, TSP & K-OPT
20. /hij1nx/level-replicator ! fork it & roll your own