You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Zhu Zhu (Jira)" <ji...@apache.org> on 2022/06/28 03:56:00 UTC
[jira] [Created] (FLINK-28271) Add configuration options for speculative scheduler
|  |
[Zhu Zhu](https://issues.apache.org/jira/secure/ViewProfile.jspa?name=zhuzh)
**created** an issue
---|---
|
---
| [Flink](https://issues.apache.org/jira/browse/FLINK) / [](https://issues.apache.org/jira/browse/FLINK-28271)
[FLINK-28271](https://issues.apache.org/jira/browse/FLINK-28271)
---
[Add configuration options for speculative
scheduler](https://issues.apache.org/jira/browse/FLINK-28271)
| Issue Type: |  Sub-task
---|---
Assignee: | Unassigned
Components: | Runtime / Configuration
Created: | 28/Jun/22 03:55
Fix Versions: | 1.16.0
Priority: |  Major
Reporter: | [Zhu
Zhu](https://issues.apache.org/jira/secure/ViewProfile.jspa?name=zhuzh)
|
Following config options are needed for speculative scheduler:
* jobmanager.adaptive-batch-scheduler.speculative.enabled, default to "false". It controls whether to enable speculative execution. Note that speculative execution only works with AdaptiveBatchScheduler, which requires "jobmanager.scheduler" to be set to "AdaptiveBatch".
* jobmanager.adaptive-batch-scheduler.speculative.max-concurrent-executions, default to "2". It controls how many executions (including the original one and speculative ones) of an ExecutionVertex can execute at the same time.
* jobmanager.adaptive-batch-scheduler.speculative.block-slow-node-duration, default to "1 min". It controls how long an identified slow node should be blocked for.
---
| | [ 
](https://issues.apache.org/jira/browse/FLINK-28271#add-comment "Add Comment")
| [Add Comment](https://issues.apache.org/jira/browse/FLINK-28271#add-comment
"Add Comment")
---|---
| This message was sent by Atlassian Jira (v8.20.10#820010-sha1:ace47f9) | |

---