| |
|
EVENT DETAILS |
Details
This event is also published via https://www.meetup.com/bay-area-apache-flink-meetup/events/288746042/.
We'd like to invite you to the first event of this new meetup group, gracefully hosted by Wikimedia with talks by ThousandEyes & Immerok.
Schedule:
5:30 Social
6:00 - 6:30 Near Real-Time Statistical Modeling & Anomaly Detection using Flink, Kunal Umrigar, Engineer Leader @ ThusandEyes (part of Cisco)
6:40 - 7:10 Flink 1.16 in 30 minutes, Konstantin Knauf, Apache Flink PMC, Co-Founder @ Immerok
7:10 - ? Social
Talk Details
Near Real-Time Statistical Modeling & Anomaly Detection using Flink
Abstract: At ThousandEyes we receive billions of events every day that allow us to monitor the internet; the most important aspect of our platform is to detect outages & anomalies that have a potential to cause serious impact to customer applications & user experience. Automatic detection of such events at lowest latency & highest accuracy is extremely important for our customers & their business. After launching several resilient & low latency data pipelines in production using Flink we decided to take it up a notch; we leveraged Flink to build statistical models in near real-time & apply them on incoming stream of events to detect anomalies! In this session we will deep dive into the design as well as discuss pitfalls & learnings while developing our real-time platform that leverages Debezium, Kafka, Flink, ElasticCache & DynamoDB to process events at scale!
Bio: Kunal Umrigar is an Engineering Leader with more than 14 years of experience designing big data platforms & distributed systems. He is a Sr. Engineering Manager at ThousandEyes, leads the alerts team to build stream processing pipelines that identify network anomalies in real-time. Prior to this, Kunal worked as Sr. Director, Engineering at PubMatic to build a big data platfom that processed petabytes of data to power & optimize ad serving algorithms. He is a data & ML enthusiast always thrilled to solve tough problems, build a high-performance team & deliver quality software.
Apache Flink v1.16 in 30 minutes
Abstract: What do Apple, Netflix, ByteDance, & other digital leaders have in common? They provide a superior customer experience enabled by data-driven, real-time decisions at scale. For years, Apache Flink has been at the heart of this development, powering use cases at the intersection of real-time analytics & event-driven applications: from instant pricing & fraud detection to real-time data pipelines.
Over time, Apache Flink has become a massive project with a wide variety of use cases & initiatives: unified stream & batch processing, cloud native elasticity, PyFlink & Flink ML to name some of these.
In this talk, I will introduce the main improvements in the latest release of Apache Flink & put them into the context of those larger multi-release threads.
Bio: Konstantin is a member of the Apache Flink PMC & long-term contributor to the project. He recently co-founded Immerok with a group of long-term community members to provide the best experience for building, running, & supporting stream processing applications in the cloud. Formerly, as Head of Product at Ververica, Konstantin supported multiple teams working on Apache Flink in both discovery as well as delivery. Before that he was leading the pre-sales team at Ververica, helping their clients as well as the Open Source Community to get the most out of Apache Flink.
|
|
|
|
|
|
