Building distributed systems is accessible. I promise.

Understanding and building distributed systems can be a daunting task, but like most other software development patterns, distributed systems mimic concepts in the real world that you’re already familiar with. Jamie Winsor walks you through building a mental model to help you understand the basics of building distributed systems based on concrete, real-world systems, equipping you with the ability to digest the academic papers that form the basis of distributed systems as we know it.

Not every software development technique is a perfect fit for every problem, and distributed programming is no different. Jamie outlines the pros and cons of building distributed programs to demonstrate when you may or may not want to use this technique. As you explore these trade-offs, you’ll learn how understanding distributed systems helps you find solutions to some of your unanswered problems as well as new solutions to problems you may have previously seen as solved. Jamie discusses his own experience in how rethinking a problem with an eye for distributed systems turned some of his career’s work on its head and gave him a new passion for an old and tiresome problem.

Transcript

1. Building Distributed Systems Is Accessible, I Promise
2. EverQuest
3. Other Distributed System Examples • Distributed Databases • Blockchain • Telephone Networks • Literally The Internet
4. Daunting At First But Accessible I Promise
5. White Papers Aren't Scary I Promise
6. Time, Clocks, and The Ordering of Events in a Distributed System
7. A collection of distinct processes which are spatially separated, and which communicate with one another by exchanging messages. — Leslie Lamport
8. Distributed Processes Are Concurrent
9. A Universal Modular ACTOR Formalism for Artificial Intelligence
10. Distributed Processes Lack A Global Clock
11. Lamport Clock Sender time = time + 1; send(msg, time); Receiver (message, time_stmap) = recv(); time = max(time_stamp, time) + 1;
12. Distributed Processes Are Independent Of Failure
13. Architectures of a Distributed System
14. Client-Server
15. n-Tier
16. Peer-to-peer
17. Primitives To Build Standard Library's Net
18. What Are Distributed Systems Good At?
19. Brewer's conjecture and the feasibility of consistent, available, partition-tolerant web services
20. CAP Theorem
21. Consistency Availability Partition Tolerance
22. Applying Distributed Systems To IT Automation
23. Gossip Protocol / Epidemic Protocol
24. The Fist Bump Protocol Message
25. You Are a Peer-to-peer Network
26. Detecting Failures and Identifying Peers
27. SWIM
28. (Scalable)WIM
29. S(Weakly-consistent)IM
30. SW(Infection-style)M
31. SWI(Membership Protocol)
32. News Cast Spreading Rumors
33. Service Discovery
34. Automatic Reconfigure On Failure
35. Sub Protocols
36. Elections in A Distributed Computing System
37. Leader Follower Failover
38. Rolling Updates
39. Papers • Time, Clocks, and The Ordering of Events in a Distributed System • A Universal Modular ACTOR Formalism for Artificial Intelligence • A Robust and Scalable Peer-to-Peer Gossiping Protocol • SWIM: Scalable Weakly-consistent Infection-style Process Group Membership Protocol • Elections in A Distributed Computing System
40. Jamie Winsor @resetexistence github.com/reset