THERE WILL BE NO FOOD (for the body). 6:15 doors, talk starts at 7, out by 8:30. Thanks Target for hosting!
Title: Show me the problem - ES insights at Netflix
Every second thousands of Netflix members hit the play button to stream content on more than 1000 different Netflix device types. The playback experience is comprised of multiple dimensions - device, customer, network, content, country, languages. At Netflix’s scale, these dimensions are beyond what a human can comprehend. To enable understanding of what’s working and not working across the globe for Netflix customers, we need to blend humans and machines. Elasticsearch and Kibana augment our human analysis and enable automated root causing of streaming problems.
In this talk, we’ll discuss:
- The dimensions of data used in generating an optimal playback experience
- How we use Elasticsearch and Kibana to slice and dice these dimensions for our insights
- The automation workflow we have in place for finding the issues auto-magically and identifying the root cause of the issue.
Suudhan Rangarajan works on the Playback Features team at Netflix, responsible for ensuring that customers receive the best possible viewing experience every time they click play. His services fill a key role in enabling Netflix to stream amazing content to 69M+ members on 1000+ devices. Prior to Netflix, Suudhan worked on the Audio/Video decoding pipeline in Adobe Flash and Adobe Primetime products helping many partners create a great video streaming client. Suudhan enjoys developing and running large scale distributed services, with special focus on insights, debuggability and efficiency. Suudhan has received a Masters in Computer Science from University of Texas at Austin.