Map & Territory: A story of visibility

1. Map & Territory a story of visibility
2. Pierre-Yves @pyr https://github.com/pyr
3. https://exoscale.ch
4. Visibility
5. How do we work ?
7. How do we improve?
8. Avoid Shortcuts!
10. We want lower defect rates
11. We want to make informed decisions
12. Design Build Live
13. Visibility
14. Extracting meaningful state data from heterogeneous event sources, over time
15. Meaningful (relates to business value)
16. State Data (structured payload)
17. Heterogeneous (everyone is involved)
18. Over time (tracking)
19. How does it help my system's lifecycle ?
20. Map =/= Territory
21. Break out of our mental model
22. "I'll push this minor change, it cannot do any harm"
24. "I'll just add this static route"
26. Better lifecycle Informed decisions Better maps
27. Systems are (increasingly) complex
28. Web Infrastructure circa 00 (2 servers)
29. Visibility Circa '00
30. Web Infrastructure circa '12 (27 nodes)
32. Visibility Circa '12
33. Q: how is business doing today ? A:
34. Q: how is business doing today ? A: based on these key metrics we're looking good
35. Figure out those key metrics
36. We need appropriate tooling
37. events across: system, components, software
38. The event stream approach
39. Plenty of small producers Few big consumers
40. Production: Anything that happens or moves (logs too!): Normalize & Stream
41. Consumption: Aggregate Correlate Decide
42. Aggregation compute compound metrics (ratios, sums)
43. Correlation
44. Decision track, alert, ignore, scale
45. Implementing on premise, saas or in between ?
46. SaaS loggly, papertrail, librato, datadog, ...
47. On Premise collectd, logstash, graphite, statsd, riemann
48. The path to visibility: Find key metrics Find the right tools Rely on an event stream Involve everyone Challenge your mental model Hopefully, improve quality and lower defect rates in the process!
49. Questions ?