[
https://issues.apache.org/jira/browse/BEAM-135?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15946946#comment-15946946
]
Etienne Chauchot commented on BEAM-135:
---------------------------------------
The orientation of this feature is to provide exact batching
(https://docs.google.com/document/d/1zf9TxIOsZf_fz86TGaiAQqdNI5OO7Sc6qFsxZlBAMiA/edit?usp=sharing)
to allow use cases such as abuse detection in an auction system: raise an
alert when a user has made 10 auctions in the last hour. The state API allows
to count to 10 even if a bundle is smaller than 10 elements. There are still
corner cases in which, the exact batching cannot be obtained: if the window
contains 15 elements, then there will be a batch of 10 elements and a batch of
5 elements because we need to output the batched elements in the same window as
the input.
> Utilities for "batching" elements in a DoFn
> -------------------------------------------
>
> Key: BEAM-135
> URL: https://issues.apache.org/jira/browse/BEAM-135
> Project: Beam
> Issue Type: New Feature
> Components: sdk-java-core
> Reporter: Ben Chambers
> Assignee: Etienne Chauchot
>
> We regularly receive questions about how to write a {{DoFn}} that operates on
> batches of elements. Example answers include:
> http://stackoverflow.com/questions/35065109/can-datastore-input-in-google-dataflow-pipeline-be-processed-in-a-batch-of-n-ent/35068341#35068341
> http://stackoverflow.com/questions/30177812/partition-data-coming-from-csv-so-i-can-process-larger-patches-rather-then-indiv/30178170#30178170
> Possible APIs could be to wrap a {{DoFn}} and include a batch size, or to
> create a utility like {{Filter}}, {{Partition}}, etc. that takes a
> {{SerializableFunction}} or a {{SimpleFunction}}.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
