You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Roman Khachatryan (Jira)" <ji...@apache.org> on 2020/08/18 17:06:00 UTC

[jira] [Updated] (FLINK-18991) Optimize reading InputChannel state

     [ https://issues.apache.org/jira/browse/FLINK-18991?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Roman Khachatryan updated FLINK-18991:
--------------------------------------
    Description: 
Curently, channel state is read not sequentially.

Inverting control would make it more efficient.

Current call chain: 
{code:java}
 StreamTask.readRecoveredChannelState ->  strictly after reading output; when state is *consumed*, requests partitions
    SingleInputGate.readRecoveredState -> submits a task to an executor from main task thread; loop through special channels
        RecoveredInputChannel.readRecoveredState  -> loop while has data; buffer = bufferManager.requestBufferBlocking();
                where bufferManager = new BufferManager(inputGate.getMemorySegmentProvider(), this, 0);
            ChannelStateReader.readInputData  
{code}
Proposed call chain:
{code:java}
StreamTask.readRecoveredChannelState ->  strictly after reading output; when state is *consumed*, requests partitions
    ChannelStateReader.readInputData (another thread, maybe same executor, maybe change injection)
        loop through state handles ordererd by handle, offset
        request buffer from channel
        pass buffer to RecoveredInputChannel.onRecoveredStateBuffer
{code}
  Depends on FLINK-18989

  was:
Curently, channel state is read not sequentially.

Inverting control would make it more efficient.

Current call chain: 
{code:java}
 StreamTask.readRecoveredChannelState ->  strictly after reading output; when state is *consumed*, requests partitions
    SingleInputGate.readRecoveredState -> submits a task to an executor from main task thread; loop through special channels
        RecoveredInputChannel.readRecoveredState  -> loop while has data; buffer = bufferManager.requestBufferBlocking();
                where bufferManager = new BufferManager(inputGate.getMemorySegmentProvider(), this, 0);
            ChannelStateReader.readInputData  
{code}
Proposed call chain:
{code:java}
StreamTask.readRecoveredChannelState ->  strictly after reading output; when state is *consumed*, requests partitionsStreamTask.readRecoveredChannelState ->  strictly after reading output; when state is *consumed*, requests partitions    ChannelStateReader.readInputData (another thread, maybe same executor, maybe change injection)        loop through state handles ordererd by handle, offset        request buffer from channel        pass buffer to RecoveredInputChannel.onRecoveredStateBuffer
{code}
  Depends on FLINK-18989


> Optimize reading InputChannel state
> -----------------------------------
>
>                 Key: FLINK-18991
>                 URL: https://issues.apache.org/jira/browse/FLINK-18991
>             Project: Flink
>          Issue Type: Improvement
>          Components: Runtime / Task
>            Reporter: Roman Khachatryan
>            Assignee: Roman Khachatryan
>            Priority: Major
>             Fix For: 1.12.0
>
>
> Curently, channel state is read not sequentially.
> Inverting control would make it more efficient.
> Current call chain: 
> {code:java}
>  StreamTask.readRecoveredChannelState ->  strictly after reading output; when state is *consumed*, requests partitions
>     SingleInputGate.readRecoveredState -> submits a task to an executor from main task thread; loop through special channels
>         RecoveredInputChannel.readRecoveredState  -> loop while has data; buffer = bufferManager.requestBufferBlocking();
>                 where bufferManager = new BufferManager(inputGate.getMemorySegmentProvider(), this, 0);
>             ChannelStateReader.readInputData  
> {code}
> Proposed call chain:
> {code:java}
> StreamTask.readRecoveredChannelState ->  strictly after reading output; when state is *consumed*, requests partitions
>     ChannelStateReader.readInputData (another thread, maybe same executor, maybe change injection)
>         loop through state handles ordererd by handle, offset
>         request buffer from channel
>         pass buffer to RecoveredInputChannel.onRecoveredStateBuffer
> {code}
>   Depends on FLINK-18989



--
This message was sent by Atlassian Jira
(v8.3.4#803005)