The Fork in the API by Ivan Dwyer

The Fork in the API by Ivan Dwyer

Transcript

  1. The Fork in the API App-centric vs Job-centric Microservices Ivan

    Dwyer | Head of Business Development | Iron.io | @fortyfivan
  2. “Loosely coupled service oriented architecture with bounded contexts” - Adrian

    Cockroft characteristics of a microservice
  3. a containerized workflow sets the stage run ship build Dockerfile

    Docker Image Docker Container Docker Daemon ?
  4. “The intended lifecycle of the workload itself determines how a

    microservice should behave” - Me but what about its runtime behavior?
  5. a further distinction needs to be made App-centric Real-time requests

    Job-centric Background processes
  6. how does it run? Synchronous Asynchronous request 1 response 1

    request 2 response 2 wait wait request 1 response 1 request 2 response 2
  7. how is it deployed? Staged Uploaded Runtime Image Repo push

    upload
  8. how is it invoked? Request Event request Client Event trigger

    response
  9. how is it routed? Load Balancer Queue Load Balancer

  10. how does it fail? Reroute Retry Load Balancer retry

  11. how does it scale? Elastic Concurrent Load Balancer

  12. how does it execute? Long Running Ephemeral start stop process

  13. how does it get data? Connection Payload DB

  14. the API Gateway brings it all together Load Balancer Message

    Queue API Gateway webhook API request API request schedule app job
  15. within a complete cloud native stack Infrastructure Storage Compute Networking

    Platform 12-Factor Apps Configuration Orchestration Microservices App Services Data Services Job Services API Gateway Security & Authentication Monitoring & Logging
  16. within a (in)complete vendor stack Infrastructure Platform Microservices API Gateway

    Security & Authentication Monitoring & Logging
  17. event-driven computing for cloud native applications queue, schedule and execute

    job-centric microservices at scale www.iron.io