Transformations
Transformations are the basic building block of Cloud Dataflow pipelines, with each transformation representing a step of the overall processing task. Developers define each transformation by implementing PTransform, which accepts one or more PCollection, operates on the elements within that collection, and returns zero or more PCollection as a result of those operations.
By largely abstracting away the underlying bounded/unbounded nature of the data, PCollection allow developers to focus on the PTransform algorithm while allowing Cloud Dataflow to intelligently handle sharding, parallelization, and dynamic rebalancing. Each PTransform can be categorized into one of three types: element-wise transforms, aggregate transforms, ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access