Pesquisa Viva: Distributed Systems Patterns

Method

#

(General rules: see .claude/skills/research/SKILL.md.)

• Organized by pattern, not by series part. Each pattern gathers the matching part of krishnadaspc’s series and the related Scholion notes.
• The series builds each pattern from spawn, send and receive, without GenServer, GenStage or Broadway. When a direction reaches the OTP abstraction that formalizes the pattern, keep the two levels apart: what the series builds by hand vs. what the library provides.
• Most related notes are webclips (secondary summaries). Claims taken from them need the original article before getting ✓.

Estado

#

• Em foco: none yet; the author will pick the first direction later.
• Próximo: choose which direction opens the research.

Motivation

#

Starting point: the series Building Distributed Systems in Elixir, by krishnadaspc (pckrishnadas88) on dev.to, published between 2026-07-30 and 2026-09-22. Part 10 (Connecting Nodes) is announced but not yet published.

Links: Part 9 — Backpressure · author on dev.to

Open questions

#

Directions / Readings

#

(To be confirmed with the author. Each one is exhausted before moving to the next.)

1. State in a process

#

Part 1: Building a Stateful Process in Elixir Without GenServer (2026-07-30). Scholion webclip: building-a-stateful-process-in-elixir-without-genserver. Related: erlang, elixir, otp-open-telecom-platform, thinking-in-actors-part-1, jonas-boner-actor-model-akka-reactive-programming-microservi. ?

2. Request–reply and service communication

#

Part 2: Correlated Request–Reply (2026-08-06). Scholion webclip: building-distributed-systems-with-elixir-02-correlated-reque. Related: must-know-service-communication-patterns, cross-app-communication-with-rpc-in-elixir, retries-have-an-evil-twin-duplicates. ?

3. Failure detection

#

Part 3: Process Monitoring (2026-08-13). Part 4: Process Linking (2026-08-14). Scholion webclips: building-distributed-systems-in-elixir-part-3-process-monito, building-distributed-systems-in-elixir-part-4-process-linkin. Related: heartbeats-in-distributed-systems. ?

4. Supervision and resiliency

#

Part 5: Supervisor From Scratch (2026-08-18). Scholion webclip: building-distributed-systems-in-elixir-part-5-supervisor-fro. Related: must-known-resiliency-patterns-for-distributed-systems, building-robust-distributed-systems, http-best-practices-using-asp-net-core-and-polly. ?

5. Naming and discovery

#

Part 6: Named Processes (2026-08-19). Scholion webclip: building-distributed-systems-in-elixir-part-6-named-processe. ?

6. Worker pools

#

Part 7: Worker Pool (2026-08-24). Scholion webclip: building-distributed-systems-in-elixir-part-7-worker-pool. Related: dont-ignore-thread-safety-design-for-concurrency-from-day-on. ?

7. Publish/subscribe and events

#

Part 8: Publish / Subscribe (2026-09-14). Scholion webclip: building-distributed-systems-in-elixir-part-8-publish-subscr. Related: distributed-pubsub-in-elixir, understanding-concepts-in-event-driven-architectures, the-dual-nature-of-events-in-event-driven-architecture, patterns-best-practices-event-driven-systems, serverless-event-driven-systems. ?

8. Backpressure

#

Part 9: Backpressure (2026-09-22). ⚠ From a first read, not yet checked line by line: three experiments, going from uncoordinated push (the consumer’s mailbox grows without bound), to stop-and-wait with acknowledgements, to a consumer-driven demand window; mentions GenStage and Broadway as the production form of the demand protocol. Scholion webclip: building-distributed-systems-in-elixir-part-9-backpressure. Related: elixir-makes-you-make-good-decisions, notes-on-distributed-systems-for-young-bloods, high-performance-dotnet-apps-with-csharp-channels. ?

9. Clustering and consistency

#

Part 10 (Connecting Nodes), announced. Related: cap-theorem-for-databases-consistency-availability-partition, designing-data-intensive-applications-book, rick-reed-whatsapp-scaling-erlang. ?

Related Scholion notes

#

General overviews, not tied to a single direction:

• 8-must-know-distributed-system-design-patterns — patterns for failure handling, read/write separation, partitioning and coordination.
• 15-must-know-elements-of-system-design — system design elements grouped by area.
• system-design-tips — caches, queues, sharding and replication, with when to use each.
• review-is-designing-data-intensive-applications-worth-it — review of Kleppmann’s book.
• death-by-a-thousand-microservices — the cost of distributing without need.

Notes tied to a direction are listed inside it above.

Extracted notes

#