Webinar: Apache Apex Fault Tolerance & Processing Semantics
1Apache Apex (http://apex.incubator.apache.org/) is an open source stream processing and next generation analytics platform incubating at the Apache Software Foundation. Apex is Hadoop native and was built from ground up for scalability, low-latency processing, high availability and operability.
In this webinar, you will learn about Apache Apex fault tolerance, high availability and processing guarantees.
From the users perspective, fault tolerance of a stream processing platform should cover the state of the application/processor and the in-flight data. In the event of failure, the platform should recover, restore state and resume processing with no loss of data. We will cover:
* Components of an Apex application and how they are made fault tolerant
* How native YARN support is leveraged for fault tolerance
* How operator checkpointing works and how the user can tune it
* Failure scenarios, recovery from failures, incremental recovery
* Processing guarantees and which option is appropriate for your application
* Sample topology for highly available, low latency real-time processing
* How is fault-tolerance in Apex different from similar platforms such as Storm, Spark Streaming and Flink.
Speakers: Thomas Weise PPMC member & Co-Founder, Architect at DataTorrent. Pramod Immaneni PPMC member & Architect at DataTorrent.