Microservices at Gigya

Slides from the first meetup about Microdot, Gigya's open source microservices framework in .NET.

Transcript

1. By Meetup #1 - Introduction
2. Microservices @
3. About Gigya • Customer Identity Management Platform • Multiple products • Hundreds of (big enterprise) customers • 1.3 billion user profile managed • 5 Data centers • 10K API calls/sec (with X10 peaks)
4. We started small
5. A couple of years later…
6. We had problems • One code base, multiple teams • Archeological code layers • Technical debt (not enough unit tests, weird hacks) • “All or nothing” deployment • Looooooooong release cycles
8. Microservices Benefits • Scalability (organizational and technical) • Agility • Resiliency • And also… (easier to understand, easier to deploy, easier to change, easier to experiment, multiple technologies)
9. Microservices – Not a free lunch! Modularity System Complexity
10. Microservices – Not a free lunch! • System complexity • Operational complexity • Performance and latency • Reliability • Consistency • Debugability
12. And we had another problem
13. Concurrency
14. Distributed Concurrency
16. (A is for Actors)
17. The Actor Model • Simpler concurrent programming model • Write single-threaded code (easier to understand) • Concurrency and scale via actor instances • Maximizes CPU utilization • Easy to distribute
18. Virtual Actors • Started in 2008 at Microsoft Research – Project Orleans • In production since 2011 (Halo 4 cloud services) • Open sourced and moved to Microsoft Studios at 2015 • A port to Java – Orbit (by EA)
19. Virtual Actors • Goals: • Make distributed application programming easier • Prefer developer productivity and transparent scalability • “A programming model and runtime for building cloud native services”
20. Virtual Actors • A Virtual Actor: always exists and never fails
21. Virtual Actors Actor types: • Worker • An auto-scaling processing unit – multiple instances created by framework as needed • Single Activation • Guaranteed to have a single active instance in the cluster • A Stateful application middle-tier
22. Virtual Actor Framework • A runtime providing virtual “actor space”, analogues to virtual memory • Handles Actor placement, activation and GC when needed • Balances resources across the cluster, provides elastic scalability
23. Simplified Programming Model • An Actor is a class, implementing an interface with asynchronous methods • The caller of an Actor uses the actor interface via a proxy • Messaging is transparent and handled by the runtime. Programmers deal with interfaces and methods
24. Simplified Programming Model
25. Simplified Programming Model
26. So can we use Actors as microservices?
27. A Microservice • “Small” (developed by a single team) • Solves a problem (single responsibility) • Interacts with other services over well-defined interfaces • Independently deployed • Remains consistent and available in presence of failures
28. Actors as Microservices ? • “Small” (developed by a single team) • Solves a problem (single responsibility) • Interacts with other services over well-defined interfaces • Independently deployed • Remains consistent and available in presence of failures
29. Actors as Microservices ? • “Small” (developed by a single team) • Solves a problem (single responsibility) • Interacts with other services over well-defined interfaces • Independently deployed • Remains consistent and available in presence of failures (?)
30. Actor based microservices Service A Service B ServiceB Interface Actor IMyServiceB
31. Actor based microservices Service A Service B ServiceB Interface Actor IMyServiceB ServiceB HTTP Listener
32. Actor based microservices Service A Service B ServiceB Interface Actor IMyServiceB ServiceB Client ServiceB HTTP Listener JSON over HTTP