Stream: Chapel
Time: 12:15 - 13:00
Mainframe applications have been processing high-value business transactions in real time for decades. As enterprises increasingly adopt event-driven architectures, cloud-native platforms, and AI-powered applications, a new challenge emerges: how do you continuously process and act on data as it flows across the enterprise? Apache Flink is an open-source distributed stream processing platform designed to analyse and transform data in motion. It enables developers to build applications that process events continuously, maintain state across millions of transactions, and deliver low-latency insights at massive scale. This session introduces Apache Flink from a mainframe practitioner's perspective. We will explore Flink's programming model, state management, fault tolerance, and event-time processing capabilities, drawing parallels with familiar z/OS concepts such as transactional integrity, workload management, and high availability. We will also examine how Flink complements existing mainframe environments by enabling real-time integration between systems of record on z/OS and modern cloud-based applications. Attendees will leave with a practical understanding of where Apache Flink fits within an enterprise architecture, the problems it solves, and why it is becoming a key technology for real-time data processing.
There is currently no attachment for From transactions to streams: understanding Apache Flink in enterprise computing
Andrew is the architect for z/OS Connect focussing on providing the ability for z/OS customers to unlock the data in their systems through modern RESTful APIs. He is also involved with open source maintaining several projects related to z/OS Connect and contributing to the Zowe project.
Click here to give some Feedback so we can make it even better next year!