Scaling a Distributed Stateful System: A LinkedIn Case Study
summary
Covers how LinkedIn scaled a distributed stateful data system by addressing replication, data distribution, load balancing, cluster management, client data handling, capacity management, and tuning. Describes techniques for moving and replicating data within hot clusters to avoid degrading service performance.
AI-written summary · about
chapters (4)
related talks
02h 18mDistributed Systems: In the Daily Life of the Data Engineer
26mScaling LinkedIn’s search infrastructure
41mIntroduction to distributed systems
34mScaling Applications with MongoDB: Best Practices and Challenges
27mScaling for Intelligence: Distributed SQL Databases in the Age of AI
47mA Distributed, Scalable and Highly Reliable Data Storage System
45mHow LinkedIn is leveraging kafka for log aggregation and monitoring
30mDistributed system design