Using Spark at Vungle

This is a past event

180 people went

Location image of event venue


This meetup will be at Vungle and feature talks by Vungle engineers about their use of Spark. Find a detailed talk abstract and speaker bios below.

Agenda for the evening:
6:30 - 7:00 :: Mingling
7:00 - 7:05 :: Welcome and Intro to Vungle
7:05 - 8:30 :: Technical talk
8:30 - 9:00 :: Mingling

The talks will be live streamed and the video will be published on the Apache Spark channel ( on YouTube.

Also Spark Summit is two weeks later (June 15/16). We have a great set of talks from NASA, CIA, Netflix, Baidu, Airbnb, Microsoft, and more. Use the discount code “SFmeetup” to get 15% off registration.


Talk Abstract:

Vungle enables developers to put video ads in their apps. In this talk, we will discuss Spark's critical role in Vungle's infrastructure. Spark's flexibility and power allowed us to easily horizontally scale a variety of systems within our stack to power over 1 billion events per day. From completely replacing existing data processing systems to reinforcing stressed components, Spark's versatility, intuitive development model, and ease of deployment have made it a very attractive choice for a number of data processing problems we face at Vungle.

== Speaker Bios ==

Tony Chong - Devops & Data Engineering Manager:

Tony is a self starter and highly motivated DevOps leader. For the last several years his main focus has been forming and implementing DevOps practices that allow businesses requirements to scale rapidly into technology solutions. With a proven track record and several awards, Tony's recent interests and projects have been in containerization, orchestration and service discovery.

Ben Reiter - Data Engineer

Ben has been working at startups for the past three and a half years, working mostly on various data processing pipelines. Initially he was involved with migrating a batch oriented Map Reduce pipeline to EMR, and advocated additional use of the pipeline more to allow for horizontally scaling infrastructure. At Vungle, he has been part of the team that built out a brand new data pipeline that uses Flume, Kafka, Spark and Cassandra.

Sasan Hezarkhani - Server Engineer

Sasan is a Software Engineer at Vungle working on the ad serving team. Previously he was at Retargeter helping with building a reporting infrastructure using Spark and SparkSQL. Before that he was working at EA on the Origin team.

== Parking info ==

- 450 5th Street
- 325 5th Street
- 777 Harrison Street (closest)