Leveraging ETS Effectively

on YouTube ↗46m

" summary

Deep dive into ETS internals with concrete claims, code and benchmarks: match specifications are shown as the real power, and a demo replaces an Agent that chokes at ~16k concurrent items with ETS that handles ~500k reads/updates; practical patterns include using ETS as presets (static data), as sharded caches per scheduler, as message buffers and lightweight counters (with an Atomics comparison), and an Ecto adapter that maps schemas to ETS for queries and streaming (repo: github.com/evadne/ets-playground).

AI-written summary · about

" chapters (16)

1. 0:00Start
2. 0:43self Elixir Enthusiast. Bug Farmer Badger of Applications
3. 2:09Conceptual Model 2
4. 3:17Table Types
5. 3:29Access Types
6. 3:48Key Functions
7. 4:49ets:match_object/2
8. 8:48Demo: Agent vs. ETS Perhaps a demonstration will help you make the right decision for your application
9. 11:56Mental Model: Complexity Gradient
10. 13:15Sharing Preset Data
11. 15:49With Presets: Easy to Scale
12. 17:00Storing Cached State
13. 18:48Facilitating IPC
14. 19:44Concurrency Control
15. 23:48Adopting ETS: Cookie Cutter Layout
16. 25:14Adopting ETS: With Ecto

" related talks

• 52mOn the Scalability of the Erlang Virtual Machine and its Term StorageKostis Sagonas · Erlang Factory SF Bay Area 2014
• 37mLeveling Up With EctoDarin Wilson · ElixirConf 2016
• 41mRevisiting XTM: A Practical Case Study Highlighting Its NeedsYuya Watari · PGConf.dev 2025
• 35mElixir's Ecto: Functional Relational Data Access Done RightBrad Urani · LambdaConf 2018
• 41mReal-World Datomic: An Experience ReportCraig Andera · Strange Loop 2013
• 44mElixir and moneyTomasz Kowal · Lambda Days 2017
• 39mEcto without SQLGuilherme de Maio · ElixirConf EU 2019
• 33mExponential Growth Without Sacrificing Platform SpeedGil Allouche, Prithvi Sathiya · Aerospike User Summit 2018