Flink Forward Virtual Conference 2020
Stream Processing | Event Driven | Real Time
About Flink Forward
Flink Forward is the conference for Apache Flink® and the stream processing communities. Join core Flink committers, new and experienced users, and thought leaders to share experiences and best practices in stream processing, real-time analytics, event-driven applications, and managing mission-critical Flink deployments in production.

Hear from experts who have built scalable streaming infrastructure and enterprise-grade applications, view live demos and join interactive in-depth technical talks on the internals of stream processing technologies. Participants learn how global companies like Google, Yelp, Alibaba, Netflix, Lyft use Flink as the stream processing engine of choice for large-scale stateful applications, including real-time analytics, real-time search and content ranking, fraud/anomaly/threat detection.

Flink Forward Virtual Conference kicks off on April 22 with live keynotes and talks on production Flink use cases, to Apache Flink® internals, to the growth of the Flink ecosystem, and many more topics on stream processing and real-time analytics.
    • April 22: 8:30 am - 9:20 am PDT
      • Keynote: Introducing Stateful Functions 2.0: Stream Processing meets Serverless Applications
        We are presenting the new version of Stateful Functions (2.0), the first version released as part of the Apache Flink project.

        Stateful Functions was designed to apply the ideas of stream processing to event-driven applications. Version 2.0 brings big changes, most prominently a new architecture that allows you to pull the computation out of the Flink processes and run them efficiently as separate services. It means you can run your application's logic on FaaS platforms (like AWS lambda, AliCloud Function Compute, etc.) or as Kubernetes deployments. You can operate the application like a stateless service (rapid scaling, rolling upgrades) while maintaining the consistency guarantees from Apache Flink's stream processing APIs. And, as a nice extra, it is language-agnostic, so it opens the ecosystem up to many other programming languages. We believe it forms a great foundation for combining the strengths of stream processing and serverless technology, making way for a Stateful Serverless platform.
      • Stephan Ewen
        CTO & Co-founder at Ververica

        Stephan Ewen is CTO and co-founder at Ververica where he leads the development of the stream processing platform based on... View More
    • April 22: 9:20 am - 9:40 am PDT
      • Keynote: The Easiest Way to Get Operational With Apache Flink
        Over the last years, we at Ververica have supported numerous companies - digitals as well as traditional enterprises - to adopt Apache Flink as their central stream processing infrastructure. Aside from our continuous contributions to the Open Source Community, our enterprise offering, Ververica Platform, has played a central role in this process.

        In this talk, you will learn how Ververica Platform can build the foundation of your company's journey with Apache Flink as well as how our recently released, free Community Edition makes getting started with Apache Flink easier than ever.
    • April 22: 9:50 am - 10:30 am PDT
      • Keynote: Apache Flink - Completing Cloudera’s End to End Streaming Platform
      • Marton Balassi
        Marton is a Flink PMC member and one of the first contributors to the streaming API. He has driven big data adoption at around 50 customers as a Senior... View More
      • Joe Witt
        Joe Witt is Vice President of Engineering at Cloudera focused on the Cloudera Data Flow (CDF) product. Joe spent 10 years at NSA most of which developing... View More
    • April 22
      • Room 1
      • Room 2
      • Room 3
      • Room 4
      • Room 5
    • April 22: 10:40 am - 11:20 am PDT
      • Apache Flink in Production with Ververica Platform
      • Everything is connected: How watermarking, scaling, and exactly once impact one another in Pravega
        Speakers: Flavio Paiva Junqueira, Tom Kaitchuck

        Company: Dell EMC

        Stream storage enables ingestion and permanent storage of continuously generated... View More
      • Netflix Data Mesh: Composable Data Processing
        Speaker: Justin Cunningham

        Company: Netflix

        Netflix processes trillions of events and petabytes of data a day in the Keystone data pipeline, which is... View More
      • Managing Real-time Data Teams
        Speaker: Jesse Anderson

        Company: Big Data Institue

        When a team moves into real-time big data, they tend to focus on the technical challenges ahead of... View More
      • Real time identity graph
        Speaker: Fakrudeen Ali Ahmed

        Company: Adobe

        Identity services team at Adobe builds identity graph that links multiple identities for a user across... View More
    • April 22: 11:30 am - 11:50 am PDT
      • Flink acceptance testing and state compatibility checking
        Speaker: Catlyn Kong

        Company: Yelp

        Since the introduction of Flink at Yelp in 2017, tens of real-time Flink applications have been developed to drive a... View More
      • Needle in the Haystack - Monitoring health of a huge Kafka fleet with Flink
        Speakers: Tirtha Chatterjee, Ryan Nienhuis

        Company: AWS

        Amazon Managed Streaming for Apache Kafka (MSK) is a fully managed, highly available, and... View More
      • Reliable, Real-Time AI with the Ververica Platform + a Kubernetes Operator in a Growing Startup
        Speaker: Austin Cawley-Edwards

        Company: Fintech Studios

        FinTech Studios uses the Ververica Platform to drive our Kubernetes-native Continuous Delivery... View More
      • Cogynt: Flink without code
        Speakers: Cogynt: Samantha Chan, Aslam Tajwala (Q&A support)

        Company: Cogility Software

        In our data-driven world, the need for speed has never been... View More
      • Tuning Flink Filesystem Plugins for State Persistence
        Speaker: Scott Kidder

        Company: Mux

        Flink can persist checkpoint and savepoint state on numerous fault-tolerant filesystems, including HDFS, S3, and... View More
    • April 22: 12:00 pm - 12:40 pm PDT
      • Flink SQL in 2020: Time to show off!
        Speakers: Fabian Hueske, Timo Walther

        Company: Ververica

        Four years ago, the Apache Flink community started adding SQL support to ease and unify the... View More
      • Large-scale near-real-time (NRT) data analytics platform empowered by Apache Flink
        Speakers: Ying Xu, Kailash Hassan Dayanand

        Company: Lyft

        In this talk, we present Lyft's streaming data analytics platform, which powers near real-time... View More
      • Writing an interactive streaming SQL and materialized view engine using Flink
        Speaker: Kenny Gorman

        Company: Eventador.io

        In this talk I will cover our journey building an interactive SQL interface and materialized view engine... View More
      • Autoscaling Flink at Netflix
        Speaker: Timothy Farkas

        Company: Netflix

        Keystone Data Pipeline manages several thousand Flink pipelines, with variable workloads. These pipelines are... View More
      • Customizing exactly-once processing at scale with Flink
        Speaker: Andrew Torson

        Company: Salesforce

        Flink offers several options for exactly-once processing guarantee: all require support from the underlying... View More
    • Day 2
      • April 23 2020
    • April 23: 8:30 am - 9:10 am PDT
      • Keynote: The Evolution of Data Infrastructure at Splunk
        Over the past few years almost all data processing has moved from batch to stream processing. This isn’t simply driven by a desire for lower latency, but by a fundamental understanding that streams are a more effective primitive for data processing, providing a better impedance match to varied downstream systems and services. Splunk, like many others, has been evolving its core data infrastructure to better provide a simpler and more consistent programming model, address correctness and latency of data, and allow for a more open integration model with our data platform. Throughout this process, we’ve come to view Apache Flink as a critical backbone in our core data infrastructure. Join us to learn more about how our data infrastructure - and how we think about it - has fundamentally changed.
      • Speaker: Eric Sammer, Distinguished Engineer at Splunk
        Bio: Eric was an early employee at Cloudera before he founded Rocana which was acquired by Splunk. Today, he is a Distinguished Engineer at Splunk working on... View More
    • April 23: 9:20 am - 10:00 am PDT
      • Keynote: Stream analytics made real with Pravega and Apache Flink
        Streaming applications have evolved and the systems that support such applications along with them. Applications today are part of complex ecosystems and not standalone applications. As such, the expectation of the features offered by streaming systems and platforms are more stringent than ever. With such requirements in mind, we have designed and built Pravega, a software system that exposes stream as a core storage primitive. Pravega enables applications to ingest and store unbounded data streams in an elastic and consistent manner. Combining Pravega and Apache Flink, we are able to build data pipelines that are able to tail and historically process stream data, to dynamically and to satisfying exactly-once semantics end-to-end. We have joined Pravega and Apache Flink in the Streaming Data Platform, a Dell Technologies product, and we share in this keynote our experience so far with the users of the platform. Use cases span from analyzing drone video streams to computer vision in connected cars.
      • Speaker: Srikanth Satya, VP of Engineering at Dell EMC
        Bio: Srikanth Satya is Vice President of Engineering for Unstructured Data Storage, responsible for Object Storage product (ECS) and is also building a... View More
    • April 23 2020
      • Room 1
      • Room2
      • Room 3
      • Room 4
    • April 23: 10:10 am - 10:50 am PDT
      • Apache Flink Worst Practices
        Speaker: Konstantin Knauf

        Company: Ververica

        Distributed stream processing is evolving from a technology in the sidelines of Big Data to a key enabler... View More
      • High-Quality Performant and Cost Efficient Schema-Aware Data Streams on Flink at Netflix Scale
        Speakers: Jagannathrao Mudda, Ramayan Tiwari

        Company: Netflix

        Over 4 million events per second with hundreds of event types and a petabyte of data per... View More
      • Production-Ready Flink and Hive Integration - what story you can tell now?
        Speaker: Bowen Li

        Company: Alibaba

        In this talk we will go thru the full Flink-Hive integration released in Flink 1.10. I will throw a few common use... View More
      • Lessons learned on Apache Flink application availability in a hosted Apache Flink service
        Speakers: Praveen Gattu, Hwanju Kim, Ryan Nienhuis

        Company: AWS

        AWS provides customers a fully managed service for running Apache Flink applications... View More
    • April 23: 11:00 am - 11:40 am PDT
      • Geospatial Situation Detection through FlinkCEP @Uber
        Speaker: Teng (Niel) Hu

        Company: Uber

        Uber operates in the complex physical world. One of the challenges of providing a reliable service is detecting... View More
      • Stateful Functions: Polyglot Event-Driven Functions for Stateful Distributed Applications
        Speakers: Tzu-Li (Gordon) Tai, Igal Shilman (Q&A support), Stephan Ewen (Q&A support)

        Company: Ververica

        Frameworks like Kubernetes have solved most of... View More
      • Energize Multi-tenancy Flink on K8s with YuniKorn
        Speakers: Weiwei Yang, Wilfred Spiegelenburg

        Company: Cloudera

        When we attempt to run large scale Flink jobs on K8s, we observed lots of problems... View More
      • Data Warehouse, Data Lakes, What's Next?
        Speaker: Xiaowei Jiang,

        Company: Alibaba

        Flink is used to build real time machine learning pipelines at Alibaba. I will go over the architecture... View More
    • April 23: 11:50 am - 12:10 pm PDT
      • Building metric platform using Flink for massive scale at Netflix
        Speaker: Abhay Amin

        Company: Netflix

        Over 150+ million members worldwide are enjoying TV series, feature films across a wide variety of genres and... View More
      • Equality and inclusion as a way of developing diverse, resilient, and more relevant OSS
        Speakers: Griselda Cuevas, Markos Sfikas

        Companies: Google, Ververica

        Have you ever wished that your bug list would decrease at a faster speed? Have... View More
      • Building FeatureFlow, Criteo’s feature data generation platform
        Speaker: Piyush Narang

        Company: Criteo

        Criteo is one of the leading advertising platforms for the open Internet. We serve over a billion users across... View More
      • Building Stateful Streaming Pipelines
        Speaker: Ankit Jhalaria

        Company: GoDaddy

        Walk attendees through our journey of building a streaming data platform at godaddy using apache beam and... View More
    • April 23: 12:20 pm - 1:00 pm PDT
      • Distributed Processing for Machine Learning Production Pipelines
        Speakers: Robert Crowe, Reza Rokni, Ahmet Altay

        Company: Google

        Production ML workloads often require very large compute and system resources, which... View More
      • State Unlocked
        Speakers: Seth Wiesman, Tzu-Li (Gordon) Tai

        Company: Ververica

        As stateful streaming processing becomes more and more mature for complex event-driven... View More
      • How Lyft built a streaming data platform with Flink on Kubernetes
        Speaker: Micah Wylde

        Company: Lyft

        Access to real-time data is increasingly important for many organizations. This is particularly true for Lyft, which... View More
      • Build your next-generation stream platform based on Apache Pulsar
        Speaker: Neng Lu

        Company: StreamNative

        As organizations are getting better at capturing streaming data and the data velocity and volume are... View More
    • April 24
      • Room 1
      • Room 2
      • Room 3
      • Room 4
    • April 24: 9:00 am - 9:40 am CEST
      • Testing production streaming applications
        Speakers: Gyula Fora, Matyas Orhidi

        Company: Cloudera

        One of the biggest challenge in streaming application development is making sure your pipeline... View More
      • AI made easy with Flink + AIFlow
        Speaker: Becket Qin

        Company: Alibaba

        There have been a lot of projects helping the users build their AI platforms, such as mlflow, tfx, metaflow,... View More
      • Rethinking of fault tolerance in Flink: what lies ahead?
        Speakers: Piotr Nowojski, Yuan Mei

        Companies: Ververica, Alibaba

        Flink is state of the art streaming processing framework providing exactly-once... View More
      • A deep dive into Flink SQL
        Speaker: Kurt Young, Jark Wu

        Company: Alibaba

        During last two major versions (1.9 & 1.10), Apache Flink community spent lots of effort to improve the... View More
      • Room 1
      • Room 2
      • Room 3
    • April 24: 9:50 am - 10:10 am CEST
      • Extend Flink to edge computing with much lower footprint
        Speaker: Youjun Yuan

        Company: Baidu

        Running Flink on resource constrained edge devices is hard. There are two major obstacles: a) It consumes too much... View More
      • Data driven matchmaking streaming at Hyperconnect
        Speakers: Jaehyeuk Oh, Gihoon Yeom

        Company: HyperConnect

        HyperConnect's 1to1 video matchmaking system is consist of various machine learning techniques... View More
      • Machine learning with Flink in Weibo
        Speaker: Yu Qian

        Company: Weibo

        This session will talk about how we use Flink to do real-time processing and online machine learning in Weibo... View More
    • April 24: 10:20 am - 11:00 am CEST
      • Integrate Flink with Kubernetes natively
        Speaker: Yang Wang

        Company: Alibaba

        Currently Flink supports the resource management system YARN and Mesos. However, they were not designed for fast... View More
      • How Streaming Helps Your Staging Environment and Sandboxes Stay Up To Date
        Speaker: Mike Solomon

        Company: Meeshkan

        In this session, we will look at how Apache Flink can be used to stream anonymized API request and response... View More
      • It’s finally here: Python on Flink & Flink on Zeppelin
        Speakers: Jincheng Sun, Jeff Zhang

        Company: Alibaba

        Flink has made tremendous progress in its core engine to unify batch processing and streaming... View More
Space Is Limited: Reserve Your Spot Today!