Playback data systems
- 1. Distributed Systems Engineer Playback Data Systems https://jobs.netflix.com/jobs/861178 Ketan Duvedi
- 2. Netflix The world’s leading Internet TV network 190+ countries86+ million members 1500+ devices supported 125 million hours / day
- 3. High scale
- 4. Big Global Opportunity Global broadband households continue to grow
- 5. Innovative Culture Values are what we value High Performance Freedom & Responsibility Context, not Control Highly Aligned, Loosely Coupled Pay top of market Promotions & Development http://www.slideshare.net/reed2001/culture-1798664
- 6. Team Focus Build and Operate server side infrastructure to collect, process, and provide high quality data about Netflix playback experiences.
- 7. Data Collection Viewing Data Playback events Player QoE Errors Ingest > 500K events/sec
- 8. Data Collection Device Attributes Firmware, SDK, UI State changes Ingest ~ 50K events/sec
- 10. Data Processing - High Volume @ Low Latency PDS owns 4 of Top 5 Cassandra Clusters at Netflix
- 11. Data Providing What you’ve watched Where you’re at Real time Operational Insights
- 12. Data Providing Data feeds ● Events and summaries ● Large number of consumers Personalized Recommendations
- 13. DevOps Culture Insights into our systems enables Operational Excellence You Build It, You Run It
- 14. Technology Stack
- 16. Global Scale Improve the microservices architecture to ensure linear scalability at sublinear cost.
- 17. Data quality Deliver high quality data despite diversity of quality in input events. Use data to deliver business functionality
- 18. Flexible Processing Build platform to enable partners to quickly explore new data transformations and views.
- 19. For more info... Details of the team’s work and approaches: ● How Netflix knows where you are in House of Cards ● Viewing History System Evolution
Editor's Notes
- #2: Background image from https://www.flickr.com/photos/theholyllama/4143228359/, no changes made, https://creativecommons.org/licenses/by-sa/2.0/
- #4: Data for North America, Fixed Access networks, from Sandvine’s biannual reports.
- #6: Image from https://www.flickr.com/photos/ericsbardo/3757385525/, no changes made, https://creativecommons.org/licenses/by-sa/2.0/
- #7: The team acts as the “data edge” for playback-oriented interactions.
- #8: How much was watched What was the quality What’s changing on the device over time
- #9: How much was watched What was the quality What’s changing on the device over time
- #10: Sessionize Sanitize / clean Synthesize Summarize Aggregate
- #19: Through approaches like containerized stream processing rather than existing solutions that are skewed towards batch processing.
- #21: Apply at https://jobs.netflix.com/jobs/861178 or contact us at [email protected] or [email protected]