[jira] [Updated] (FLINK-28823) Enlarge the max requested buffers for SortMergeResultPartitionReadScheduler

2022-08-07 Thread Yingjie Cao (Jira)


 [ 
https://issues.apache.org/jira/browse/FLINK-28823?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yingjie Cao updated FLINK-28823:

Fix Version/s: 1.16.0

> Enlarge the max requested buffers for SortMergeResultPartitionReadScheduler
> ---
>
> Key: FLINK-28823
> URL: https://issues.apache.org/jira/browse/FLINK-28823
> Project: Flink
>  Issue Type: Sub-task
>  Components: Runtime / Network
>Affects Versions: 1.16.0
>Reporter: Yuxin Tan
>Assignee: Yuxin Tan
>Priority: Major
>  Labels: pull-request-available
> Fix For: 1.16.0
>
>
> The num of the max requested buffers in SortMergeResultPartitionReadScheduler 
> is determined by the num of subpartitions, which may be insufficient in some 
> scenarios, such as speculative execution and partition reuse.
> Therefore, we should increase the max requested buffers to make full use of 
> available memory when TM memory is sufficient.
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)


[jira] [Updated] (FLINK-28823) Enlarge the max requested buffers for SortMergeResultPartitionReadScheduler

2022-08-05 Thread ASF GitHub Bot (Jira)


 [ 
https://issues.apache.org/jira/browse/FLINK-28823?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

ASF GitHub Bot updated FLINK-28823:
---
Labels: pull-request-available  (was: )

> Enlarge the max requested buffers for SortMergeResultPartitionReadScheduler
> ---
>
> Key: FLINK-28823
> URL: https://issues.apache.org/jira/browse/FLINK-28823
> Project: Flink
>  Issue Type: Sub-task
>  Components: Runtime / Network
>Affects Versions: 1.16.0
>Reporter: Yuxin Tan
>Priority: Major
>  Labels: pull-request-available
>
> The num of the max requested buffers in SortMergeResultPartitionReadScheduler 
> is determined by the num of subpartitions, which may be insufficient in some 
> scenarios, such as speculative execution and partition reuse.
> Therefore, we should increase the max requested buffers to make full use of 
> available memory when TM memory is sufficient.
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)