[
https://issues.apache.org/jira/browse/BEAM-7131?focusedWorklogId=247341&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-247341
]
ASF GitHub Bot logged work on BEAM-7131:
----------------------------------------
Author: ASF GitHub Bot
Created on: 23/May/19 08:33
Start Date: 23/May/19 08:33
Worklog Time Spent: 10m
Work Description: robertwb commented on pull request #8558: [BEAM-7131]
Spark: cache executable stage output to prevent re-computation
URL: https://github.com/apache/beam/pull/8558#discussion_r286830540
##########
File path:
runners/spark/src/main/java/org/apache/beam/runners/spark/translation/SparkBatchPortablePipelineTranslator.java
##########
@@ -111,6 +112,25 @@ public void translate(final RunnerApi.Pipeline pipeline,
SparkTranslationContext
QueryablePipeline p =
QueryablePipeline.forTransforms(
pipeline.getRootTransformIdsList(), pipeline.getComponents());
+ for (PipelineNode.PTransformNode transformNode :
p.getTopologicallyOrderedTransforms()) {
+ // Pre-scan pipeline to count which pCollections are consumed as inputs
more than once so
+ // their corresponding RDDs can later be cached.
+ for (String inputId :
transformNode.getTransform().getInputsMap().values()) {
+ context.incrementConsumptionCountBy(inputId, 1);
+ }
+ // Executable stage consists of two parts: computation and extraction.
This means the result
+ // of computation is an intermediate RDD, which we might also need to
cache.
Review comment:
Is this preferable to extracting the RDD corresponding to the (potentially
expensive) extraction? (In that case, rather than special casing
ExecutableStage, we could simply increment all outputs of multi-output
transforms.)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 247341)
> Spark portable runner appears to be repeating work (in TFX example)
> -------------------------------------------------------------------
>
> Key: BEAM-7131
> URL: https://issues.apache.org/jira/browse/BEAM-7131
> Project: Beam
> Issue Type: Bug
> Components: runner-spark
> Reporter: Kyle Weaver
> Assignee: Kyle Weaver
> Priority: Major
> Time Spent: 4h 40m
> Remaining Estimate: 0h
>
> I've been trying to run the TFX Chicago taxi example [1] on the Spark
> portable runner. TFDV works fine, but the preprocess step
> (preprocess_flink.sh [2]) fails with the following error:
> RuntimeError: AlreadyExistsError: file already exists [while running
> 'WriteTransformFn/WriteTransformFn']
> Assets are being written multiple times to different temp directories, which
> is okay, but the error occurs when they are copied to the same permanent
> output directory. Specifically, the copy tree operation in transform_fn_io.py
> [3] is run twice with the same output directory. The error doesn't occur when
> that code is modified to allow overwriting existing files, but that's only a
> shallow fix. While the TF transform should probably be made idempotent, this
> is also an issue with the Spark runner, which shouldn't be repeating work
> like this regularly (in the absence of a failure condition).
> [1] [https://github.com/tensorflow/tfx/tree/master/tfx/examples/chicago_taxi]
> [2]
> [https://github.com/tensorflow/tfx/blob/master/tfx/examples/chicago_taxi/preprocess_flink.sh]
> [3]
> [https://github.com/tensorflow/transform/blob/master/tensorflow_transform/beam/tft_beam_io/transform_fn_io.py#L33-L45]
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
