Evolution of the Transcoder Architecture
summary
Describes a video transcoding system that converts user videos into popular resolutions, computes video fingerprints, and generates subtitles. Explains how the system evolved to prioritize live (interactive) users versus batch processing jobs and the approaches used to balance and schedule those workloads.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 2:59CPU utilization issues
- 8:16System requirements and loads
- 10:48Solution: smart scheduler
- 15:30Scheduler architecture
- 23:59Results and future plans
- 24:43Q&A
related talks
27mSummarizing & analyzing videos
21mWhat to do after per-title encoding
22mMovie Gen Video: State-of-the-art Video Generation
40mSmart Crop and Smart Preview via Video Understanding
33mMaximize lifetime value of media content
45mTranscend Language Barriers with Custom Machine Learning APIs: um...
16mSwift: Adaptive Video Streaming with Layered Neural Codecs
20mReconstructing proprietary video streaming algorithms
