Skip to main content

23. Tumbling Fixed Event-Time WindowingHard

Windowing & Streaming⏱️ ~20 mins

23. Tumbling Fixed Event-Time Windowing

Enterprise Architecture Context

In production stream-processing architectures (Google Cloud Dataflow / Flink), pipeline stages must handle parallel transformations without data loss, managing schema mutations and aggregations across distributed worker workers.

Problem Statement

### Business Context Operational metrics dashboards aggregate server HTTP requests into discrete 10-second non-overlapping tumbling time windows. ### Problem Statement Write a function `window_ten(input_pcoll)` that assigns streaming elements into fixed event-time windows of duration **10 seconds** using `beam.WindowInto(FixedWindows(10))`.

Key Learning Objectives

  • Understand distributed Apache Beam execution DAG stages and pipeline lifecycle.
  • Apply idiomatic functional Python transforms using the pipe operator |.
  • Ensure data consistency and idempotency across distributed stream workers.

Sample Data Fixtures

Sample Example 1
Input Stream:
[10, 20, 30]
Expected Output:
[10, 20, 30]
Sample Example 2
Input Stream:
[5]
Expected Output:
[5]
Topics:#Windowing#Streaming
Support