You have got to love the transparency of how Netflix deliver their product from a tech perspective. The latest iteration of Netflix Open Source Software Meetup which is completely open to the public took place this week in Los Gatos featuring eight new projects speakers with lightening talks and community users from IBM Watson, Nike Digital and Pivotal.
Netflix do not have to do this but it’s a bit like opening the hood of a secretive military machine to the public to show how they make the magic work and this week’s event featured talks from
- Roy Raport from in house telemetry platform Atlas talks about the sweet horror of creating their own solution when there are so many other OS and off the shelf products in the market and what they have in store. (go to 0:35)
- Diptanu Choudhury from the platform engineering team spoke about the platform features and how they have learnt about the behaviour of the platform and improved it with load balancing, monitoring libraries and Microservices and how Prana copes with all the different languages and tools used in the Netflix ecosystem (go to 6:30)
- Sagar Loke is a Cloud and Platform Engineer who talks about Raigad which runs as a co-process alongside Elastic Search helping to automate deployment (go to 10:55)
- Tom Gianos from the big data platform team spoke about scalable REST-ful execution and configuration services for big data jobs (go to 15:55)
- Daniel Weeks is from the same team and speaks about Inviso which curates configuration information about jobs across distributed systems (go to 20:37)
- Minh Do spoke about the Netflix tool Dynamite entitled “Zero To Docker: An easy way to evaluate NetflixOSS” which delves into configuration layers (go to 24:40)
- Vasanth Asokan talks about Nicobar, a dynamic scripting library and module loading framework for Java which enables UI customisation from Netflix’s ‘experience based APIs’ (go to 29:12)
- Mitch Zollinger talks about how and why the Message Security Layer (MSL) enables arbitrary application protocols to be secured over transport protocols, performance, reliability and protocol design (go to 33:42)
Related Posts
