[
https://issues.apache.org/jira/browse/FLINK-7416?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16211297#comment-16211297
]
ASF GitHub Bot commented on FLINK-7416:
---------------------------------------
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4533#discussion_r145750939
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/CreditBasedClientHandler.java
---
@@ -37,18 +43,29 @@
import java.io.IOException;
import java.net.SocketAddress;
+import java.util.ArrayDeque;
import java.util.concurrent.ConcurrentHashMap;
import java.util.concurrent.ConcurrentMap;
import java.util.concurrent.atomic.AtomicReference;
+/**
+ * Channel handler to read {@link BufferResponse} and {@link
ErrorResponse} messages from the
+ * producer, to write and flush {@link AddCredit} message for the producer.
+ */
class CreditBasedClientHandler extends ChannelInboundHandlerAdapter {
private static final Logger LOG =
LoggerFactory.getLogger(CreditBasedClientHandler.class);
+ /** Channels, which already requested partitions from the producers. */
--- End diff --
In this PR I add the new `inputChannelsWithCredit` queue and comment it. In
order to contrast
with previous `inputChannels`, I also add comment for it. As you said, it
can also be done in last commit.
> Implement Netty receiver outgoing pipeline for credit-based
> -----------------------------------------------------------
>
> Key: FLINK-7416
> URL: https://issues.apache.org/jira/browse/FLINK-7416
> Project: Flink
> Issue Type: Sub-task
> Components: Network
> Reporter: zhijiang
> Assignee: zhijiang
> Fix For: 1.4.0
>
>
> This is a part of work for credit-based network flow control.
> The related works are :
> * We define a new message called {{AddCredit}} to notify the incremental
> credit during data shuffle.
> * Whenever an {{InputChannel}}’s unannounced credit goes up from zero, the
> channel is enqueued in the pipeline.
> * Whenever the channel becomes writable, it takes the next {{InputChannel}}
> and sends its unannounced credit. The credit is reset to zero after each sent.
> * That way, messages are sent as often as the network has capacity and
> contain as much credit as available for the channel at that point in time.
> Otherwise, it would only add latency to the announcements and not increase
> throughput.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
