This is an automated email from the ASF dual-hosted git repository.
gyfora pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/flink-kubernetes-operator.git
The following commit(s) were added to refs/heads/main by this push:
new 2d4c992f [hotfix][docs] Document deprecation of
target.utilization.boundary in favor of utilization.max / utilization.min
2d4c992f is described below
commit 2d4c992f5ff48d9a2e7cb1404a08ceaa5a9c803e
Author: Dennis-Mircea Ciupitu <[email protected]>
AuthorDate: Tue Apr 7 12:19:22 2026 +0300
[hotfix][docs] Document deprecation of target.utilization.boundary in favor
of utilization.max / utilization.min
---
docs/content.zh/docs/custom-resource/autoscaler.md | 16 +++++++++++++---
docs/content/docs/custom-resource/autoscaler.md | 16 +++++++++++++---
2 files changed, 26 insertions(+), 6 deletions(-)
diff --git a/docs/content.zh/docs/custom-resource/autoscaler.md
b/docs/content.zh/docs/custom-resource/autoscaler.md
index b70eee91..c1763494 100644
--- a/docs/content.zh/docs/custom-resource/autoscaler.md
+++ b/docs/content.zh/docs/custom-resource/autoscaler.md
@@ -156,8 +156,17 @@ A target of `0.6` means we are targeting 60%
utilization/load for the job vertex
In general, it's not recommended to set target utilization close to 100% as
performance usually degrades as we reach capacity limits in most real world
systems.
-In addition to the utilization target we can set a utilization boundary, that
serves as extra buffer to avoid immediate scaling on load fluctuations.
-Setting `job.autoscaler.target.utilization.boundary: "0.2"` means that we
allow 20% deviation from the target utilization before triggering a scaling
action.
+To control when scaling actions are triggered, users can set upper and lower
utilization thresholds using `job.autoscaler.utilization.max` and
`job.autoscaler.utilization.min`.
+These define the utilization range within which the autoscaler will not take
any scaling action, providing a buffer against load fluctuations.
+
+For example, with a target utilization of `0.6`, setting
`job.autoscaler.utilization.max: "0.8"` and `job.autoscaler.utilization.min:
"0.4"` means:
+- Scale-up is triggered when utilization exceeds `0.8`
+- Scale-down is triggered when utilization drops below `0.4`
+- No scaling action is taken while utilization remains between `0.4` and `0.8`
+
+{{<hint info >}}
+The default thresholds are `target + 0.3` for max and `target - 0.3` for min.
+{{< /hint >}}
### Target catch-up duration and restart time
@@ -183,7 +192,8 @@ flinkConfiguration:
job.autoscaler.stabilization.interval: 1m
job.autoscaler.metrics.window: 5m
job.autoscaler.target.utilization: "0.6"
- job.autoscaler.target.utilization.boundary: "0.2"
+ job.autoscaler.utilization.max: "0.8"
+ job.autoscaler.utilization.min: "0.4"
job.autoscaler.restart.time: 2m
job.autoscaler.catch-up.duration: 5m
pipeline.max-parallelism: "720"
diff --git a/docs/content/docs/custom-resource/autoscaler.md
b/docs/content/docs/custom-resource/autoscaler.md
index 0b3ee757..8c7fc2e7 100644
--- a/docs/content/docs/custom-resource/autoscaler.md
+++ b/docs/content/docs/custom-resource/autoscaler.md
@@ -156,8 +156,17 @@ A target of `0.6` means we are targeting 60%
utilization/load for the job vertex
In general, it's not recommended to set target utilization close to 100% as
performance usually degrades as we reach capacity limits in most real world
systems.
-In addition to the utilization target we can set a utilization boundary, that
serves as extra buffer to avoid immediate scaling on load fluctuations.
-Setting `job.autoscaler.target.utilization.boundary: "0.2"` means that we
allow 20% deviation from the target utilization before triggering a scaling
action.
+To control when scaling actions are triggered, users can set upper and lower
utilization thresholds using `job.autoscaler.utilization.max` and
`job.autoscaler.utilization.min`.
+These define the utilization range within which the autoscaler will not take
any scaling action, providing a buffer against load fluctuations.
+
+For example, with a target utilization of `0.6`, setting
`job.autoscaler.utilization.max: "0.8"` and `job.autoscaler.utilization.min:
"0.4"` means:
+- Scale-up is triggered when utilization exceeds `0.8`
+- Scale-down is triggered when utilization drops below `0.4`
+- No scaling action is taken while utilization remains between `0.4` and `0.8`
+
+{{<hint info >}}
+The default thresholds are `target + 0.3` for max and `target - 0.3` for min.
+{{< /hint >}}
### Target catch-up duration and restart time
@@ -183,7 +192,8 @@ flinkConfiguration:
job.autoscaler.stabilization.interval: 1m
job.autoscaler.metrics.window: 5m
job.autoscaler.target.utilization: "0.6"
- job.autoscaler.target.utilization.boundary: "0.2"
+ job.autoscaler.utilization.max: "0.8"
+ job.autoscaler.utilization.min: "0.4"
job.autoscaler.restart.time: 2m
job.autoscaler.catch-up.duration: 5m
pipeline.max-parallelism: "720"
