10 Ways to Reach HighLoad and BigData Out of Thin Air
summary
The talk analyzes ten practical ways a web project can accidentally become a HighLoad/BigData case by common development choices. It names database-related issues as primary drivers and discusses concrete architectural and operational patterns that lead to high load and how they emerge during normal web development and maintenance.
AI-written summary · about
chapters (11)
- 0:00Start
- 4:51Scaling
- 6:18Scaling
- 12:32EAV simplifies design
- 13:35ORM simplifies development
- 15:55The main evil in PostgreSQL is autovacuum
- 18:48JOIN is evil — they are slow
- 20:37Let's invent Slony
- 24:04Everything works in my test
- 29:46Be smart, as a java-developer!
- 34:11Nice little things
related talks
56mHow to Create High Load Out of Thin Air
06h 53mHighLoad++ 2018 Live Stream. November 9, Congress Hall
01h 00mBurn Brightly So the Database Doesn’t Go Out, or How to Live with High Database Loads
46mTechniques for Developing a High-Load Server in Java
47mQuality Control for High-Load Systems
47mLoad Testing You Always Wanted
41mTo Big Data or Not Too Big
37mPractical Experience Building APIs for High-Load Services