Hello Apache Beam Community, First, thank you for the incredible work and vision behind this project.
My team is currently evaluating Apache Beam as the standard for our data processing platform. The primary driver for us is Beam's powerful unified API for both batch and streaming workloads, which is a key strategic goal for our architecture. As part of our evaluation, we are assessing the available runners and their future compatibility. The Apache Flink runner is a strong candidate for our execution engine. We've noted that Flink 2.0 was released in March 2025, introducing significant architectural changes like the removal of the DataSet API. I have looked through the official Beam roadmap page and searched on JIRA, but I haven't been able to find a definitive plan or timeline regarding support for Flink 2.0. To help with our evaluation, could you please provide some insight on the following? 1. What is the community's current plan or thinking regarding support for the Flink 2.0 runner? 2. Are there specific technical challenges that are the main focus for this effort? 3. Is there a primary JIRA Epic or a github tracking issue that we can follow to stay updated on the progress? Any clarity you can provide on the roadmap would be immensely helpful for us to make an informed decision on adopting Apache Beam for our unified processing needs. Thank you for your time and guidance!
