[
https://issues.apache.org/jira/browse/BEAM-14429?focusedWorklogId=767485&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-767485
]
ASF GitHub Bot logged work on BEAM-14429:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 07/May/22 01:40
Start Date: 07/May/22 01:40
Worklog Time Spent: 10m
Work Description: aaltay commented on code in PR #17576:
URL: https://github.com/apache/beam/pull/17576#discussion_r867285500
##########
.test-infra/jenkins/job_LoadTests_CoGBK_Dataflow_V2_Java11.groovy:
##########
@@ -47,6 +47,7 @@ def loadTestConfigurations = { mode, isStreaming ->
"numRecords": 20000000,
"keySizeBytes": 10,
"valueSizeBytes": 90,
+ "forceNumInitialBundles": 1,
Review Comment:
Could you add a comment with TODO(BEAM-14429) here and in other places?
Issue Time Tracking
-------------------
Worklog Id: (was: 767485)
Time Spent: 1.5h (was: 1h 20m)
> SyntheticUnboundedSource(with SDF) produce wrong number of records when
> initial split is larger than 1
> ------------------------------------------------------------------------------------------------------
>
> Key: BEAM-14429
> URL: https://issues.apache.org/jira/browse/BEAM-14429
> Project: Beam
> Issue Type: Bug
> Components: io-common
> Reporter: Yichi Zhang
> Priority: P2
> Time Spent: 1.5h
> Remaining Estimate: 0h
>
> With the default 20 split, the num records produced by
> Read.from(SyntheticUnboundedSource) is always larger than the numRecords
> specified. the more splits the more actual number records produced is off.
> And the Read step tends to take longer time with more splits.
>
> The issue is manifested with java LoadTests on dataflow runner v2.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
