[
https://issues.apache.org/jira/browse/FLINK-8458?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16334156#comment-16334156
]
ASF GitHub Bot commented on FLINK-8458:
---------------------------------------
Github user alpinegizmo commented on a diff in the pull request:
https://github.com/apache/flink/pull/5317#discussion_r162906372
--- Diff: docs/ops/config.md ---
@@ -290,6 +290,12 @@ The following parameters configure Flink's JobManager
and TaskManagers.
- `taskmanager.network.numberOfBuffers` (deprecated, replaced by the three
parameters above): The number of buffers available to the network stack. This
number determines how many streaming data exchange channels a TaskManager can
have at the same time and how well buffered the channels are. If a job is
rejected or you get a warning that the system has not enough buffers available,
increase this value (DEFAULT: **2048**). If set, it will be mapped to
`taskmanager.network.memory.min` and `taskmanager.network.memory.max` based on
`taskmanager.memory.segment-size`.
+- `taskmanager.network.memory.buffers-per-channel`: Number of network
buffers to use for each outgoing/incoming channel (subpartition/input channel).
Especially in credit-based flow control mode, it indicates how many credits are
exclusive in each input channel. It should be configured at least 2 for good
performance. 1 buffer is for receving in-flight data in the subpartition and 1
buffer is for parallel serialization.
+
+- `taskmanager.network.memory.floating-buffers-per-gate`: Number of extra
network buffers to use for each outgoing/incoming gate (result partition/input
gate). In credit-based flow control mode, it indicates how many floating
credits are shared for all the input channels. The floating buffers are
distributed based on backlog (real-time output buffers in the subpartition)
feedback. So the floating buffers can help relief back-pressure caused by
imbalance data distribution among subpartitions.
+
--- End diff --
Number of extra network buffers used by each outgoing/incoming gate (result
partition/input gate). In credit-based flow control mode, this indicates how
many floating credits are shared among all the input channels. The floating
buffers are distributed based on backlog (real-time output buffers in the
subpartition) feedback, and can help relieve back-pressure caused by unbalanced
data distribution among the subpartitions.
> Add the switch for keeping both the old mode and the new credit-based mode
> --------------------------------------------------------------------------
>
> Key: FLINK-8458
> URL: https://issues.apache.org/jira/browse/FLINK-8458
> Project: Flink
> Issue Type: Sub-task
> Components: Network
> Reporter: zhijiang
> Assignee: zhijiang
> Priority: Major
> Fix For: 1.5.0
>
>
> After the whole feature of credit-based flow control is done, we should add a
> config parameter to switch on/off the new credit-based mode. To do so, we can
> roll back to the old network mode for any expected risks.
> The parameter is defined as
> {{taskmanager.network.credit-based-flow-control.enabled}} and the default
> value is true. This switch may be removed after next release.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
