Cheaper, More Reliable, Simpler
summary
Describes building a large-scale video storage and delivery system that handles tens of petabytes and hundreds of gigabits/sec of throughput. Covers the infrastructure challenges of using thousands of disks and hundreds of servers to store Full HD/4K uploads and live streams, and discusses practical approaches to make the system cheaper, more reliable, and simpler.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 6:01Old system problems
- 10:16Correction codes solution
- 15:53Data migration
- 22:00Disk simplification
- 27:15How data is read
- 33:15Cluster management
- 35:41Implementation results
- 40:26Q&A
related talks
47mArchitecture of the National Video Hosting RUTUBE: How We Serve 5 Tbps
47mA Distributed, Scalable and Highly Reliable Data Storage System
39mHigh-volume Data Streaming and Storage
31mThe Design of a Modern Distributed Object Storage
47mWeb-Scale IT Engineering.
30mDistributed system design
28mUber HDFS Unit Storage Cost 10x Deduction
36mStreamlio: Towards a Unified End-to-end Solution