Structured Data at Box: How We're Building for Scale

on YouTube ↗23m

" summary

Covers Box’s MySQL-based sharded database and caching architecture for serving hundreds of thousands of businesses and tens of millions of users. Describes specific design choices made in that sharded setup and evaluates how those choices have worked out for Box at scale.

AI-written summary · about

" chapters (13)

1. 0:00Intro
2. 2:57Core Application Data Architecture
3. 3:48Cache usage at Box
4. 4:27Cache Pool Stats
5. 5:41Direct client access is a problem...
6. 7:55Tron Architecture
7. 9:27Relational Data at Box -Core Application Meta-Data
8. 10:40Sharding Strategy
9. 13:35Read-only Slaves
10. 14:54Challenges
11. 17:54Credence Architecture
12. 18:34Incremental Development and Deployment
13. 20:59Closing thoughts...

" related talks

• 52mTo Shard or Not to ShardPeter Zaitsev · Percona Live 2016
• 49mSharding: from scratch to Yandex DiskAndrey Kolnoochenko · HighLoad++ 2023
• 01h 00mThe Evolution of Databases at DropboxSlava Bakhmutov · HighLoad++ 2017
• 32mData @Scale Seattle- Eric Hwang introduced by Ganapathy KrishnamoorthyEric Hwang, Ganapathy Krishnamoorthy · Data @Scale Seattle 2015
• 46mDatabase Sharding the Right Way Easy, Reliable, and Open sourceEsen Sagynov, Lee Seongjun · HighLoad++ 2012
• 29mBuilding on Strengths, Learning from DifferencesBaron Schwartz · O'Reilly MySQL Conference & Expo 2011
• 27mAutomatically shard & scale-out traditional databases on K8sTrista Pan · Conf42 Cloud Native 2023
• 30mDatastore Axes: Choosing the Scalability Direction You NeedNicolai Plum · LISA18