Decouple all the things - asynchronous messaging keeps it simple

Talk at arrrrcamp 2015

Transcript

1. Decouple all the things Asynchronous messaging keeps it simple Arrrrcamp 2015
2. Communication is Exhausting
3. RSVP
4. Ambigous Reply 4 accepts X regrets Guests: all my kids Name:
5. Ambigous Reply 5 accepts X regrets X Guests: Name: Ada
6. No RSVP required?
7. Machine-to-Machine Communication
8. 418 I’m a Teapot
9. Improve Maintainability and Stability by adding Asynchronous Communication
10. Outline Disadvantages of Synchronous Communication AMQP Publish-Subscribe Task Queue Code Example & Live Demo Summary 10
11. Disadvantages of Synchronous Communication
12. inform about update Master data updates need to propagate 12 business partners master data business partner out of business orders support contracts
13. downside of REST: client misses update if temporarily down 13 master data orders support contracts
14. downside of REST: master data app needs to know its clients 14 master data orders support contracts invoices
15. xing.com – core updates have to propagate 15 core user data jobs events …
16. Asynchronous Messaging Advanced Message Queuing Protocol
17. amqp.org/about/what “The capable, commoditized, multi-vendor communications ecosystem which AMQP enables “The capable, commoditized, multi-vendor communications ecosystem which AMQP enables creates opportunities for commerce and innovation “The capable, commoditized, multi-vendor communications ecosystem which AMQP enables creates opportunities for commerce and innovation which can transform the way business is done “The capable, commoditized, multi-vendor communications ecosystem which AMQP enables creates opportunities for commerce and innovation which can transform the way business is done on the Internet “The capable, commoditized, multi-vendor communications ecosystem which AMQP enables creates opportunities for commerce and innovation which can transform the way business is done on the Internet, and in the Cloud.”
18. “open standard application layer protocol for message-oriented middleware” https://en.wikipedia.org/wiki/ Advanced_Message_Queuing_Protocol “open standard application layer protocol for message-oriented middleware”
19. Interoperability 19 message broker message consumer message producer
20. Message Flow 20 P Message Broker X Q C Producer Exchange Queue Consumer
21. Messages contain payload and routing key 21 Routing Key Payload •  application data – any form, any encoding… …and more •  Structured app-specific data (properties / attributes) •  Headers •  Annotations •  …
22. Applications are in power 22 X Q Producer Exchange Queue Consumer P C
23. No information lost if consumer down 23 X Q Producer Exchange Queue Consumer P C
24. Add consumers without touching producer 24 P X Q2 Q1 C1 C2 Producer Exchange Queue Consumer
25. Publish - Subscribe
26. Different messages for different client needs 26 master data client Owns additional data related to master data •  User account aka master data record removed Caches master data •  If master data updated or removed Mirrors master data •  Updated master data
27. Example Message routing key <...>user.deleted payload 27 { “user_id”: 42 }
28. Example Message routing key <...>profile.updated payload 28 { “user_id”: 42, “fields”: [“last_name”] }
29. Consumer subscribes, publisher does not notice 29 P X Q2 Q1 C1 C2
30. Why not include updated data? routing key <...>profile.updated payload 30 { “user_id”: 42, “fields”: [“last_name”] }
31. No guaranteed order – aim for commutativity Order of messages not guarantueed Commutativity Complement with synchronous API (e.g. REST) •  most recent data •  leading system •  message consumers follow up with REST if necessary Small payload •  easier on the broker •  just enough data for consumer to decide about making REST request 31
32. No exactly-once delivery – aim for idem-potent messages Duplicated messages •  possible e.g. if producer re-sends message after connection failure •  use idem-potent message handling 32
33. Add producers without touching the consumer 33 Producer Exchange Queue Consumer P1 P2 X Q C
34. Centralized tracking of decentralized events 34 P1 P2 X Q C event to be tracked tracking counter ++
35. Avoid n-m routing, keep breaking changes manageable 35 P1 P2 X Q2 Q1 C1 C2 P1 P2 C2 C1
36. Notifications / Events - producer state changed 36 Message Broker X Q P C event / state change
37. Task / Command – consumer state to change 37 Message Broker X Q P C event / state change
38. Task Queue
39. Run tasks asynchronously 39 Message Broker X Q P C image upload user facing image processing
40. Improve response time, avoid downtimes, gain reliability 40 Message Broker X Q P C image upload user facing image processing
41. Scaling on the fly 41 X Q C C P migration trigger data migration
42. Migration without downtime 42 Message Broker X Q P C Old system New system
43. Test without impacting users 43 Message Broker X Q P C old app user facing shadow calls to new app
44. Hello, World! Code Example & Live Demo
45. Flexible Routing via different Exchange Types Direct Exchange •  routes to all queues whose binding_key equals routing_key Fanout Exchange •  broadcasts to every queue it has a binding for •  ignores routing key and queue name Topic Exchange •  dot-separated words as routing keys, plus wildcards •  routing on multiple attributes Header Exchange •  based on headers, not routing keys •  routing on multiple attributes 45
46. In a nutshell •  Interoperability •  Improve response times •  Avoid downtimes •  Robustness & Reliability •  Scale on the fly •  Reduce complexity •  Simplify code 46
47. Decouple all your things!
48. Who am I Contact •  http://www.kpuschke.eu •  twitter: @titanoboa42 •  https://www.xing.com/profile/ Kerstin_Puschke Senior Software Engineer at XING Hamburg •  xing.com social network for business professionals •  debian, perl, javascript, ruby on rails, mysql, redis, riak, . . . We’re hiring! http://corporate.xing.com/ english/company/careers- at-xing/
49. www.xing.com Thank you for your attention!