[
https://issues.apache.org/jira/browse/SPARK-20928?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16288391#comment-16288391
]
Michael Armbrust commented on SPARK-20928:
------------------------------------------
An update on this. We've started to create subtasks break down the process of
adding this new execution mode and we are targeting an alpha version in 2.3.
The basics of the new API for Sources and Sinks (for both microbatch and
continuous mode) has been up for a few days if people want to see more details.
We'll follow with PRs to add a continuous execution engine and an
implementation of a continuous kafka connector.
Regarding some of the questions:
- The version we are targeting for 2.3 will only support map operations and
thus will not support shuffles / aggregating by windows (although the window()
operator is just a projection so will work for window assignment).
- I think the API is designed in such as way that we can build a streaming
shuffle that aligns on epochIds in the future, allowing us to easily extend the
continuous engine to handle stateful operations as well.
> SPIP: Continuous Processing Mode for Structured Streaming
> ---------------------------------------------------------
>
> Key: SPARK-20928
> URL: https://issues.apache.org/jira/browse/SPARK-20928
> Project: Spark
> Issue Type: Improvement
> Components: Structured Streaming
> Affects Versions: 2.2.0
> Reporter: Michael Armbrust
> Labels: SPIP
> Attachments: Continuous Processing in Structured Streaming Design
> Sketch.pdf
>
>
> Given the current Source API, the minimum possible latency for any record is
> bounded by the amount of time that it takes to launch a task. This
> limitation is a result of the fact that {{getBatch}} requires us to know both
> the starting and the ending offset, before any tasks are launched. In the
> worst case, the end-to-end latency is actually closer to the average batch
> time + task launching time.
> For applications where latency is more important than exactly-once output
> however, it would be useful if processing could happen continuously. This
> would allow us to achieve fully pipelined reading and writing from sources
> such as Kafka. This kind of architecture would make it possible to process
> records with end-to-end latencies on the order of 1 ms, rather than the
> 10-100ms that is possible today.
> One possible architecture here would be to change the Source API to look like
> the following rough sketch:
> {code}
> trait Epoch {
> def data: DataFrame
> /** The exclusive starting position for `data`. */
> def startOffset: Offset
> /** The inclusive ending position for `data`. Incrementally updated
> during processing, but not complete until execution of the query plan in
> `data` is finished. */
> def endOffset: Offset
> }
> def getBatch(startOffset: Option[Offset], endOffset: Option[Offset],
> limits: Limits): Epoch
> {code}
> The above would allow us to build an alternative implementation of
> {{StreamExecution}} that processes continuously with much lower latency and
> only stops processing when needing to reconfigure the stream (either due to a
> failure or a user requested change in parallelism.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
