You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@druid.apache.org by "Fly-Style (via GitHub)" <gi...@apache.org> on 2026/01/30 18:27:28 UTC
[PR] Adjust costs for burst scaleup during heavy lag for cost-based autosc… (druid)
Fly-Style opened a new pull request, #18969:
URL: https://github.com/apache/druid/pull/18969
(no comment)
--
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.
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Adjust costs for burst scaleup during heavy lag for cost-based autoscaler (druid)
Posted by "capistrant (via GitHub)" <gi...@apache.org>.
capistrant commented on code in PR #18969:
URL: https://github.com/apache/druid/pull/18969#discussion_r2748042479
##########
indexing-service/src/main/java/org/apache/druid/indexing/seekablestream/supervisor/autoscaler/CostBasedAutoScaler.java:
##########
@@ -172,6 +172,7 @@ public int computeTaskCountForScaleAction()
log.info("New task count [%d] on supervisor [%s], scaling up", taskCount, supervisorId);
} else if (!config.isScaleDownOnTaskRolloverOnly()
&& optimalTaskCount < currentTaskCount
+ && optimalTaskCount > 0
Review Comment:
good sanity check, but is it purely for safety or is it actually possible for optimalTask count to be invalid in the code as written? If yes, why isn't that having a lower bound of 1 task or something to make this impossible?
--
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.
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Adjust costs for burst scaleup during heavy lag for cost-based autoscaler (druid)
Posted by "capistrant (via GitHub)" <gi...@apache.org>.
capistrant commented on code in PR #18969:
URL: https://github.com/apache/druid/pull/18969#discussion_r2749637321
##########
indexing-service/src/main/java/org/apache/druid/indexing/seekablestream/supervisor/autoscaler/CostBasedAutoScaler.java:
##########
@@ -172,6 +172,7 @@ public int computeTaskCountForScaleAction()
log.info("New task count [%d] on supervisor [%s], scaling up", taskCount, supervisorId);
} else if (!config.isScaleDownOnTaskRolloverOnly()
&& optimalTaskCount < currentTaskCount
+ && optimalTaskCount > 0
Review Comment:
Oh lol I shoulda remembered that from the last pr I reviewed
--
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.
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Adjust costs for burst scaleup during heavy lag for cost-based autoscaler (druid)
Posted by "Fly-Style (via GitHub)" <gi...@apache.org>.
Fly-Style commented on code in PR #18969:
URL: https://github.com/apache/druid/pull/18969#discussion_r2749149767
##########
indexing-service/src/main/java/org/apache/druid/indexing/seekablestream/supervisor/autoscaler/CostBasedAutoScaler.java:
##########
@@ -172,6 +172,7 @@ public int computeTaskCountForScaleAction()
log.info("New task count [%d] on supervisor [%s], scaling up", taskCount, supervisorId);
} else if (!config.isScaleDownOnTaskRolloverOnly()
&& optimalTaskCount < currentTaskCount
+ && optimalTaskCount > 0
Review Comment:
`optimalTaskCount` may be -1 (which means no changes in task count), and we don't want to count it as successful scaledown
--
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.
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org
Re: [PR] Adjust costs for burst scaleup during heavy lag for cost-based autoscaler (druid)
Posted by "capistrant (via GitHub)" <gi...@apache.org>.
capistrant merged PR #18969:
URL: https://github.com/apache/druid/pull/18969
--
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.
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@druid.apache.org
For additional commands, e-mail: commits-help@druid.apache.org