Lessons from 1BRC: Tricks and Approaches to Performance Optimization
summary
A Java performance talk about the One Billion Row Challenge (1BRC) that walks through implementations from a simple baseline to highly optimized versions for aggregating a billion rows from a text file. It covers concrete Java coding tricks and methods that produced the largest speedups, and discusses their applicability in real projects.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:26Speaker and contest
- 3:12One BRC challenge
- 6:33Base solution profiling
- 8:00Parallel file processing
- 11:54Memory-mapped files
- 16:55Data structure optimization
- 25:41CPU specifics
- 28:54Instruction pipelining
- 31:50Bit parsing tricks
- 33:36System call tuning
- 38:36When to stop optimizing
- 39:17Payment system case study
- 44:52Summary
related talks
47mThe First 80% of Reading One Billion Rows Fast Enough
49m1BRC – Nerd Sniping the Java Community
47mConquering a Billion Rows with Modern Java Concurrency
40mHow Fast Can You Parse 1 Billion Rows in Java? – Insane Speed Test
02h 57mPushing Java to the Limits: Processing a Billion Rows in under 2 Seconds
01h 28mThe dark art of performance tuning or how to become a performance hero ...
51mSecrets of top performance in Java revealed
39mMemory Efficient Java