The Math Behind Distributed Systems
summary
Claim: representing a distributed system as a topological object lets engineers convert its connectivity into algebraic equations and prove invariants, offering a route to formal verification without brute force testing. Method: use combinatorial algebraic topology and graph theory to capture all interleavings (a frozen space of possible failures) and then reason about correctness. Status and tools: approach appears in a postgraduate thesis and papers, not yet widely deployed; industry uses observability, chaos engineering (Gremlin) and AWS examples while formal proofs remain rare and costly.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:06Passion for scale
- 4:45Math in computer science
- 7:19Distributed systems basics
- 11:00Graph theory applications
- 14:00Introducing topology
- 17:46Combinatorial topology
- 21:10Formal verification
- 25:15Mathematical proofs and summary
related talks
50mVerifying a distributed system with combinatorial topology
48mThe Verification of a Distributed System
28mReasoning about Distributed Reconfigurable Systems
11mFinding Invariants of Distributed Systems: It's a Small (Enough) World After All
38mThe Verification of a Distributed System
27mDistributed Systems, Like It or Not
28mWhy Are Distributed Systems So Hard?
52mWhat We Talk About When We Talk About Distributed Systems