Aggregations & Grouping⏱️ ~25 mins
26. Custom Associative Product Combiner
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
High-performance scientific processing computes multiplicative scale factors across distributed array partitions.
### Problem Statement
Implement a custom `beam.CombineFn` subclass `ProductCombineFn` implementing `create_accumulator`, `add_input`, `merge_accumulators`, and `extract_output` to compute the total mathematical product of all integers in a collection.
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:
[2, 3, 4]
Expected Output:
24
Sample Example 2
Input Stream:
[5, 5]
Expected Output:
25
Topics:#CombineFn#Aggregations
solution.pyPython 3.11 (Apache Beam)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Input PCollection3 elements
| # | Element / Payload |
|---|---|
| 1 | 2 |
| 2 | 3 |
| 3 | 4 |
Expected Output PCollection
24
Aggregations & Grouping⏱️ ~25 mins
26. Custom Associative Product Combiner
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
High-performance scientific processing computes multiplicative scale factors across distributed array partitions.
### Problem Statement
Implement a custom `beam.CombineFn` subclass `ProductCombineFn` implementing `create_accumulator`, `add_input`, `merge_accumulators`, and `extract_output` to compute the total mathematical product of all integers in a collection.
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:
[2, 3, 4]
Expected Output:
24
Sample Example 2
Input Stream:
[5, 5]
Expected Output:
25
Topics:#CombineFn#Aggregations