Presented By
O’Reilly + Cloudera
Make Data Work
March 25-28, 2019
San Francisco, CA

How Netflix measures app performance on 250 million unique devices across 190 countries

Vivek Pasari (Netflix), Jitender Aswani (Netflix)
11:50am12:30pm Thursday, March 28, 2019
Average rating: ***..
(3.14, 7 ratings)

Who is this presentation for?

  • Data engineers

Level

Intermediate

Prerequisite knowledge

  • A basic understanding of the big data ecosystem

What you'll learn

  • Get a glimpse into the complex data challenges of generating consistent, reliable performance metrics in a global heterogeneous environment
  • Understand the importance of unified logging and data governance
  • Learn how to process and expose large-scale highly dimensional metric datasets in Spark

Description

When its members around the world launch Netflix on their favorite device, they expect to be delighted. Great content is obviously at the core of this expectation. However, it’s increasingly important to deliver a smooth user experience while browsing the Netflix app. No one likes waiting endlessly on a loading screen.

Netflix has over 125 million members spread over 191 countries around the globe. Each day its members interact with its client applications on 250 million+ devices across various platforms, such as phones, tablets, laptops, smart TVs, game consoles, VR devices, and more. There are thousands of supported device types, each with varying levels of CPU, memory, and graphics performance. These devices are connected via mobile, wired, and WiFi networks with highly variable latency and packet loss characteristics. In such a heterogeneous environment, having access to high-quality performance data is key to building client applications which can provide a consistent quality of experience throughout the world.

Vivek Pasari dives into the data engineering and architecture behind producing actionable application performance insights. Along the way, he reveals some of the key principles, tips, and tricks for using Spark to sift through over 200 billion daily data points to produce highly dimensional performance metrics, which inform all aspects of Netflix’s end-user application development.

Photo of Vivek Pasari

Vivek Pasari

Netflix

Vivek Pasari is a senior data engineer at Netflix, where he focuses on building high-quality data assets that drive innovation in delivering highly performant, consistent, and reliable user experiences for Netflix members. Previously, he spent several years in financial technology helping build large-scale cloud and data solutions.

Photo of Jitender Aswani

Jitender Aswani

Netflix

Jitender Aswani supports the infrastructure and security data engineering teams at Netflix. His team designs, builds, and deploys scalable big data architecture and solutions to enable business and operations teams to achieve consistent capacity, reliability, and security gains. Jitender is a lifelong student of smart data products and data science solutions that push organizations to make data-inspired decisions and adopt analytics-first approaches.