[jira] [Updated] (FLINK-12078) Abstract TaskEventPublisher interface for simplifying NetworkEnvironment

[ https://issues.apache.org/jira/browse/FLINK-12078?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-12078: - Summary: Abstract TaskEventPublisher interface for simplifying NetworkEnvironment (was: Abstract

[jira] [Created] (FLINK-12078) Abstract TaskEventPublisher interface for simplify NetworkEnvironment

zhijiang created FLINK-12078: Summary: Abstract TaskEventPublisher interface for simplify NetworkEnvironment Key: FLINK-12078 URL: https://issues.apache.org/jira/browse/FLINK-12078 Project: Flink

[jira] [Commented] (FLINK-6227) Introduce the DataConsumptionException for downstream task failure

[ https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16805081#comment-16805081 ] zhijiang commented on FLINK-6227: - Thanks for the concern [~till.rohrmann]. I could work on it if needed

[jira] [Created] (FLINK-12067) Refactor the constructor of NetworkEnvironment

zhijiang created FLINK-12067: Summary: Refactor the constructor of NetworkEnvironment Key: FLINK-12067 URL: https://issues.apache.org/jira/browse/FLINK-12067 Project: Flink Issue Type: Sub-task

[jira] [Commented] (FLINK-11988) Remove legacy MockNetworkEnvironment

[ https://issues.apache.org/jira/browse/FLINK-11988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16801318#comment-16801318 ] zhijiang commented on FLINK-11988: -- Thanks [~pnowojski]， I also approval for the way of submitting PR

[jira] [Created] (FLINK-11988) Remove legacy MockNetworkEnvironment

zhijiang created FLINK-11988: Summary: Remove legacy MockNetworkEnvironment Key: FLINK-11988 URL: https://issues.apache.org/jira/browse/FLINK-11988 Project: Flink Issue Type: Task

[jira] [Created] (FLINK-11885) Introduce RecordWriterBuilder for creating RecordWriter instance

zhijiang created FLINK-11885: Summary: Introduce RecordWriterBuilder for creating RecordWriter instance Key: FLINK-11885 URL: https://issues.apache.org/jira/browse/FLINK-11885 Project: Flink

[jira] [Closed] (FLINK-5137) Config JVM options and set onto the TaskManager container

[ https://issues.apache.org/jira/browse/FLINK-5137?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-5137. --- Resolution: Won't Fix > Config JVM options and set onto the TaskManager container >

[jira] [Updated] (FLINK-11035) Notify data available to network stack immediately after finishing BufferBuilder

[ https://issues.apache.org/jira/browse/FLINK-11035?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-11035: - Affects Version/s: (was: 1.8.0) > Notify data available to network stack immediately after

[jira] [Closed] (FLINK-6095) Introduce the SlotPoolOptions to config parameters for SlotPool

[ https://issues.apache.org/jira/browse/FLINK-6095?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-6095. --- Resolution: Won't Fix > Introduce the SlotPoolOptions to config parameters for SlotPool >

[jira] [Closed] (FLINK-5703) ExecutionGraph recovery based on reconciliation with TaskManager reports

[ https://issues.apache.org/jira/browse/FLINK-5703?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-5703. --- Resolution: Won't Fix > ExecutionGraph recovery based on reconciliation with TaskManager reports >

[jira] [Closed] (FLINK-4914) Define parameters for reconciliation

[ https://issues.apache.org/jira/browse/FLINK-4914?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-4914. --- Resolution: Won't Fix > Define parameters for reconciliation > > >

[jira] [Created] (FLINK-11776) Refactor to simplify the process of scheduleOrUpdateConsumers

zhijiang created FLINK-11776: Summary: Refactor to simplify the process of scheduleOrUpdateConsumers Key: FLINK-11776 URL: https://issues.apache.org/jira/browse/FLINK-11776 Project: Flink Issue

[jira] [Commented] (FLINK-1275) Add support to compress network I/O

[ https://issues.apache.org/jira/browse/FLINK-1275?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16778988#comment-16778988 ] zhijiang commented on FLINK-1275: - [~uce], thanks for creating this valuable ticket before and I think

[jira] [Commented] (FLINK-11393) Activate default shuffle implementation and remove legacy code

[ https://issues.apache.org/jira/browse/FLINK-11393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16778842#comment-16778842 ] zhijiang commented on FLINK-11393: -- [~Salatich], I am glad to see you are interested in it and welcome

[jira] [Updated] (FLINK-11750) Replace IntermediateResultPartitionID with ResultPartitionID in ResultPartitionDeploymentDescriptor

[ https://issues.apache.org/jira/browse/FLINK-11750?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-11750: - Summary: Replace IntermediateResultPartitionID with ResultPartitionID in

[jira] [Updated] (FLINK-11750) Refactor the creation of ResultPartitionID in IntermediateResultPartition

[ https://issues.apache.org/jira/browse/FLINK-11750?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-11750: - Description: The motivation of task is for preparing the creation of {{ResultPartition}} via

[jira] [Created] (FLINK-11750) Refactor the creation of ResultPartitionID in IntermediateResultPartition

zhijiang created FLINK-11750: Summary: Refactor the creation of ResultPartitionID in IntermediateResultPartition Key: FLINK-11750 URL: https://issues.apache.org/jira/browse/FLINK-11750 Project: Flink

[jira] [Commented] (FLINK-6227) Introduce the DataConsumptionException for downstream task failure

[ https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16777549#comment-16777549 ] zhijiang commented on FLINK-6227: - The initial motivation of this task is to give a hint/indicator for

[jira] [Created] (FLINK-11726) Refactor the creation of ResultPartition and InputGate into NetworkEnvironment

zhijiang created FLINK-11726: Summary: Refactor the creation of ResultPartition and InputGate into NetworkEnvironment Key: FLINK-11726 URL: https://issues.apache.org/jira/browse/FLINK-11726 Project:

[jira] [Commented] (FLINK-6227) Introduce the DataConsumptionException for downstream task failure

[ https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16774949#comment-16774949 ] zhijiang commented on FLINK-6227: - I think this task seems a bit complicated for the beginner. First you

[jira] [Created] (FLINK-11723) Remove KvState related components from NetworkEnvironment

zhijiang created FLINK-11723: Summary: Remove KvState related components from NetworkEnvironment Key: FLINK-11723 URL: https://issues.apache.org/jira/browse/FLINK-11723 Project: Flink Issue

[jira] [Created] (FLINK-11721) Remove IOMode from NetworkEnvironment

zhijiang created FLINK-11721: Summary: Remove IOMode from NetworkEnvironment Key: FLINK-11721 URL: https://issues.apache.org/jira/browse/FLINK-11721 Project: Flink Issue Type: Sub-task

[jira] [Created] (FLINK-11707) Make InputGate interface extend AutoCloseable

zhijiang created FLINK-11707: Summary: Make InputGate interface extend AutoCloseable Key: FLINK-11707 URL: https://issues.apache.org/jira/browse/FLINK-11707 Project: Flink Issue Type: Sub-task

[jira] [Created] (FLINK-11676) Make ResultPartitionWriter extend AutoCloseable

zhijiang created FLINK-11676: Summary: Make ResultPartitionWriter extend AutoCloseable Key: FLINK-11676 URL: https://issues.apache.org/jira/browse/FLINK-11676 Project: Flink Issue Type: Sub-task

[jira] [Commented] (FLINK-6227) Introduce the DataConsumptionException for downstream task failure

[ https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16771597#comment-16771597 ] zhijiang commented on FLINK-6227: - [~Salatich], I have not worked on it yet.  And I think I have no time

[jira] [Assigned] (FLINK-6227) Introduce the DataConsumptionException for downstream task failure

[ https://issues.apache.org/jira/browse/FLINK-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang reassigned FLINK-6227: --- Assignee: (was: zhijiang) > Introduce the DataConsumptionException for downstream task failure

[jira] [Created] (FLINK-11646) Remove legacy MockRecordWriter

zhijiang created FLINK-11646: Summary: Remove legacy MockRecordWriter Key: FLINK-11646 URL: https://issues.apache.org/jira/browse/FLINK-11646 Project: Flink Issue Type: Test

[jira] [Closed] (FLINK-11441) Remove the schedule mode property from RPDD to TDD

[ https://issues.apache.org/jira/browse/FLINK-11441?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-11441. Resolution: Won't Fix > Remove the schedule mode property from RPDD to TDD >

[jira] [Created] (FLINK-11604) Extend the necessary methods in ResultPartitionWriter interface

zhijiang created FLINK-11604: Summary: Extend the necessary methods in ResultPartitionWriter interface Key: FLINK-11604 URL: https://issues.apache.org/jira/browse/FLINK-11604 Project: Flink

[jira] [Created] (FLINK-11461) Remove useless MockRecordWriter

zhijiang created FLINK-11461: Summary: Remove useless MockRecordWriter Key: FLINK-11461 URL: https://issues.apache.org/jira/browse/FLINK-11461 Project: Flink Issue Type: Task

[jira] [Created] (FLINK-11460) Remove useless AcknowledgeStreamMockEnvironment

zhijiang created FLINK-11460: Summary: Remove useless AcknowledgeStreamMockEnvironment Key: FLINK-11460 URL: https://issues.apache.org/jira/browse/FLINK-11460 Project: Flink Issue Type: Task

[jira] [Created] (FLINK-11403) Remove ResultPartitionConsumableNotifier from ResultPartition

zhijiang created FLINK-11403: Summary: Remove ResultPartitionConsumableNotifier from ResultPartition Key: FLINK-11403 URL: https://issues.apache.org/jira/browse/FLINK-11403 Project: Flink Issue

[jira] [Updated] (FLINK-10653) Introduce Pluggable Shuffle Manager Architecture

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Fix Version/s: (was: 2.0.0) 1.8.0 > Introduce Pluggable Shuffle Manager

[jira] [Commented] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16740056#comment-16740056 ] zhijiang commented on FLINK-11082: -- Consider another issue you mentioned, I think the higher CPU and

[jira] [Commented] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16739998#comment-16739998 ] zhijiang commented on FLINK-11082: -- [~pnowojski], thanks for replies! It might not make sense to

[jira] [Comment Edited] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16739246#comment-16739246 ] zhijiang edited comment on FLINK-11082 at 1/10/19 9:42 AM: --- Considering

[jira] [Commented] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16739246#comment-16739246 ] zhijiang commented on FLINK-11082: -- Considering implementation, it seems a bit complicated for both

[jira] [Created] (FLINK-11282) Merge StreamRecordWriter into RecordWriter

zhijiang created FLINK-11282: Summary: Merge StreamRecordWriter into RecordWriter Key: FLINK-11282 URL: https://issues.apache.org/jira/browse/FLINK-11282 Project: Flink Issue Type: Sub-task

[jira] [Commented] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16736665#comment-16736665 ] zhijiang commented on FLINK-11082: -- [~NicoK] yes, these two features are in parallel for Flink-1.5

[jira] [Commented] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16736663#comment-16736663 ] zhijiang commented on FLINK-11082: -- [~pnowojski], your understanding is right. It might be more obvious

[jira] [Commented] (FLINK-11035) Notify data available to network stack immediately after finishing BufferBuilder

[ https://issues.apache.org/jira/browse/FLINK-11035?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16735599#comment-16735599 ] zhijiang commented on FLINK-11035: -- Yes, I agree with your proposal. It does need enough evidence

[jira] [Commented] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16735597#comment-16735597 ] zhijiang commented on FLINK-11082: -- Thanks for replying, [~pnowojski] :) If I understood correctly,

[jira] [Commented] (FLINK-10661) Initial credit should be configured in a separate parameter

[ https://issues.apache.org/jira/browse/FLINK-10661?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16728278#comment-16728278 ] zhijiang commented on FLINK-10661: -- [~NicoK], thanks for your reply!  I should describe it more

[jira] [Updated] (FLINK-10745) Serialization and copy improvements for record writer

[ https://issues.apache.org/jira/browse/FLINK-10745?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10745: - Description: This is the umbrella issue for tracking the improvements of serialization and copy during

[jira] [Updated] (FLINK-10745) Serialization and copy improvements for record writer

[ https://issues.apache.org/jira/browse/FLINK-10745?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10745: - Summary: Serialization and copy improvements for record writer (was: Improve record writer and

[jira] [Commented] (FLINK-11035) Notify data available to network stack immediately after finishing BufferBuilder

[ https://issues.apache.org/jira/browse/FLINK-11035?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16719746#comment-16719746 ] zhijiang commented on FLINK-11035: -- [~pnowojski], thanks for your explanation! I am just not sure

[jira] [Comment Edited] (FLINK-11037) Introduce another greedy mechanism for distributing floating buffers

[ https://issues.apache.org/jira/browse/FLINK-11037?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16718422#comment-16718422 ] zhijiang edited comment on FLINK-11037 at 12/12/18 3:38 AM: [~StephanEwen]，

[jira] [Commented] (FLINK-11037) Introduce another greedy mechanism for distributing floating buffers

[ https://issues.apache.org/jira/browse/FLINK-11037?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16718422#comment-16718422 ] zhijiang commented on FLINK-11037: -- [~StephanEwen]， thanks for giving above proposals in point. :) *

[jira] [Commented] (FLINK-11037) Introduce another greedy mechanism for distributing floating buffers

[ https://issues.apache.org/jira/browse/FLINK-11037?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16714407#comment-16714407 ] zhijiang commented on FLINK-11037: -- [~till.rohrmann]， thanks for paying attention for this issue! :) *

[jira] [Commented] (FLINK-11082) Increase backlog only if it is available for consumption

[ https://issues.apache.org/jira/browse/FLINK-11082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16710966#comment-16710966 ] zhijiang commented on FLINK-11082: -- [~pnowojski], what do you think of issue? > Increase backlog only

[jira] [Created] (FLINK-11082) Increase backlog only if it is available for consumption

zhijiang created FLINK-11082: Summary: Increase backlog only if it is available for consumption Key: FLINK-11082 URL: https://issues.apache.org/jira/browse/FLINK-11082 Project: Flink Issue Type:

[jira] [Closed] (FLINK-6325) Refinement of slot reuse for task manager failure

[ https://issues.apache.org/jira/browse/FLINK-6325?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-6325. --- Resolution: Duplicate > Refinement of slot reuse for task manager failure >

[jira] [Closed] (FLINK-5138) Add the method in state backend to estimate memory by hint of state size in ResourceSpec

[ https://issues.apache.org/jira/browse/FLINK-5138?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-5138. --- Resolution: Workaround > Add the method in state backend to estimate memory by hint of state size in >

[jira] [Commented] (FLINK-6325) Refinement of slot reuse for task manager failure

[ https://issues.apache.org/jira/browse/FLINK-6325?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16708266#comment-16708266 ] zhijiang commented on FLINK-6325: - already done in FLINK-9635, close it > Refinement of slot reuse for

[jira] [Created] (FLINK-11037) Introduce another greedy mechanism for distributing floating buffers

zhijiang created FLINK-11037: Summary: Introduce another greedy mechanism for distributing floating buffers Key: FLINK-11037 URL: https://issues.apache.org/jira/browse/FLINK-11037 Project: Flink

[jira] [Assigned] (FLINK-11035) Notify data available to network stack immediately after finishing BufferBuilder

[ https://issues.apache.org/jira/browse/FLINK-11035?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang reassigned FLINK-11035: Assignee: zhijiang > Notify data available to network stack immediately after finishing >

[jira] [Commented] (FLINK-11035) Notify data available to network stack immediately after finishing BufferBuilder

[ https://issues.apache.org/jira/browse/FLINK-11035?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16704380#comment-16704380 ] zhijiang commented on FLINK-11035: -- [~pnowojski], I am not sure whether it is worth improving this

[jira] [Created] (FLINK-11035) Notify data available to network stack immediately after finishing BufferBuilder

zhijiang created FLINK-11035: Summary: Notify data available to network stack immediately after finishing BufferBuilder Key: FLINK-11035 URL: https://issues.apache.org/jira/browse/FLINK-11035 Project:

[jira] [Issue Comment Deleted] (FLINK-10966) Optimize the release blocking logic in BarrierBuffer

[ https://issues.apache.org/jira/browse/FLINK-10966?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10966: - Comment: was deleted (was: [~StephanEwen], i am interested in the actor-style stream task you

[jira] [Assigned] (FLINK-10966) Optimize the release blocking logic in BarrierBuffer

[ https://issues.apache.org/jira/browse/FLINK-10966?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang reassigned FLINK-10966: Assignee: zhijiang (was: vinoyang) > Optimize the release blocking logic in BarrierBuffer >

[jira] [Commented] (FLINK-10966) Optimize the release blocking logic in BarrierBuffer

[ https://issues.apache.org/jira/browse/FLINK-10966?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16701935#comment-16701935 ] zhijiang commented on FLINK-10966: -- [~StephanEwen], i am interested in the actor-style stream task you

[jira] [Commented] (FLINK-10966) Optimize the release blocking logic in BarrierBuffer

[ https://issues.apache.org/jira/browse/FLINK-10966?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16701931#comment-16701931 ] zhijiang commented on FLINK-10966: -- [~StephanEwen], i am interested in the actor-style stream task you

[jira] [Assigned] (FLINK-10966) Optimize the release blocking logic in BarrierBuffer

[ https://issues.apache.org/jira/browse/FLINK-10966?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang reassigned FLINK-10966: Assignee: vinoyang > Optimize the release blocking logic in BarrierBuffer >

[jira] [Assigned] (FLINK-10966) Optimize the release blocking logic in BarrierBuffer

[ https://issues.apache.org/jira/browse/FLINK-10966?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang reassigned FLINK-10966: Assignee: (was: zhijiang) > Optimize the release blocking logic in BarrierBuffer >

[jira] [Commented] (FLINK-10884) Flink on yarn TM container will be killed by nodemanager because of the exceeded physical memory.

[ https://issues.apache.org/jira/browse/FLINK-10884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16698453#comment-16698453 ] zhijiang commented on FLINK-10884: -- [~1026688210]，my above comments was just for explaining the reason

[jira] [Created] (FLINK-10995) Copy intermediate serialization results only once for broadcast mode

zhijiang created FLINK-10995: Summary: Copy intermediate serialization results only once for broadcast mode Key: FLINK-10995 URL: https://issues.apache.org/jira/browse/FLINK-10995 Project: Flink

[jira] [Updated] (FLINK-10653) Introduce Pluggable Shuffle Manager Architecture

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Summary: Introduce Pluggable Shuffle Manager Architecture (was: Fine-grained Shuffle System) >

[jira] [Updated] (FLINK-10653) Fine-grained Shuffle System

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Affects Version/s: (was: 2.0.0) > Fine-grained Shuffle System > --- > >

[jira] [Updated] (FLINK-10653) Fine-grained Shuffle System

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Description: This is the umbrella issue for improving shuffle architecture. Shuffle is the process of

[jira] [Updated] (FLINK-10653) Fine-grained Shuffle System

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Fix Version/s: 2.0.0 > Fine-grained Shuffle System > --- > >

[jira] [Updated] (FLINK-10745) Improve record writer and serializer stack

[ https://issues.apache.org/jira/browse/FLINK-10745?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10745: - Description: This is the umbrella issue for tracking the improvements of record writer and

[jira] [Commented] (FLINK-10966) Optimize the release blocking logic in BarrierBuffer

[ https://issues.apache.org/jira/browse/FLINK-10966?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16695584#comment-16695584 ] zhijiang commented on FLINK-10966: -- I have some concerns about this issue: # In credit-based flow

[jira] [Updated] (FLINK-10367) Avoid recursion stack overflow during releasing SingleInputGate

[ https://issues.apache.org/jira/browse/FLINK-10367?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10367: - Issue Type: Sub-task (was: Improvement) Parent: FLINK-7282 > Avoid recursion stack overflow

[jira] [Commented] (FLINK-10662) Refactor the ChannelSelector interface for single selected channel

[ https://issues.apache.org/jira/browse/FLINK-10662?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16694190#comment-16694190 ] zhijiang commented on FLINK-10662: -- Thanks for all the suggestions!  I think we all reach the

[jira] [Comment Edited] (FLINK-10884) Flink on yarn TM container will be killed by nodemanager because of the exceeded physical memory.

[ https://issues.apache.org/jira/browse/FLINK-10884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16694177#comment-16694177 ] zhijiang edited comment on FLINK-10884 at 11/21/18 3:47 AM: I just quickly

[jira] [Commented] (FLINK-10884) Flink on yarn TM container will be killed by nodemanager because of the exceeded physical memory.

[ https://issues.apache.org/jira/browse/FLINK-10884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16694177#comment-16694177 ] zhijiang commented on FLINK-10884: -- I just quickly reviewed the related codes. In my analysis: In the

[jira] [Commented] (FLINK-10941) Slots prematurely released which still contain unconsumed data

[ https://issues.apache.org/jira/browse/FLINK-10941?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16693079#comment-16693079 ] zhijiang commented on FLINK-10941: -- Glad to hear you are also interested in it, [~QiLuo]. We are now

[jira] [Commented] (FLINK-10941) Slots prematurely released which still contain unconsumed data

[ https://issues.apache.org/jira/browse/FLINK-10941?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16692996#comment-16692996 ] zhijiang commented on FLINK-10941: -- Hi [~QiLuo], thank you for remembering the external shuffle issue.

[jira] [Created] (FLINK-10942) Deduplicate common codes in OutputEmitterTest

zhijiang created FLINK-10942: Summary: Deduplicate common codes in OutputEmitterTest Key: FLINK-10942 URL: https://issues.apache.org/jira/browse/FLINK-10942 Project: Flink Issue Type: Sub-task

[jira] [Updated] (FLINK-10820) Simplify the RebalancePartitioner implementation

[ https://issues.apache.org/jira/browse/FLINK-10820?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10820: - Description: _The current {{RebalancePartitioner}} implementation seems a little hacky for selecting a

[jira] [Comment Edited] (FLINK-10662) Refactor the ChannelSelector interface for single selected channel

[ https://issues.apache.org/jira/browse/FLINK-10662?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16681198#comment-16681198 ] zhijiang edited comment on FLINK-10662 at 11/9/18 10:08 AM: [~pnowojski], I

[jira] [Comment Edited] (FLINK-10662) Refactor the ChannelSelector interface for single selected channel

[ https://issues.apache.org/jira/browse/FLINK-10662?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16681198#comment-16681198 ] zhijiang edited comment on FLINK-10662 at 11/9/18 10:07 AM: [~pnowojski], I

[jira] [Commented] (FLINK-10662) Refactor the ChannelSelector interface for single selected channel

[ https://issues.apache.org/jira/browse/FLINK-10662?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16681198#comment-16681198 ] zhijiang commented on FLINK-10662: -- [~pnowojski], I am currently working on this. If we change to

[jira] [Created] (FLINK-10835) Remove duplicated Round-robin ChannelSelector implementation in RecordWriterTest

zhijiang created FLINK-10835: Summary: Remove duplicated Round-robin ChannelSelector implementation in RecordWriterTest Key: FLINK-10835 URL: https://issues.apache.org/jira/browse/FLINK-10835 Project:

[jira] [Created] (FLINK-10820) Simplify the RebalancePartitioner implementation

zhijiang created FLINK-10820: Summary: Simplify the RebalancePartitioner implementation Key: FLINK-10820 URL: https://issues.apache.org/jira/browse/FLINK-10820 Project: Flink Issue Type:

[jira] [Closed] (FLINK-10321) Make the condition of broadcast partitioner simple

[ https://issues.apache.org/jira/browse/FLINK-10321?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang closed FLINK-10321. Resolution: Fixed > Make the condition of broadcast partitioner simple >

[jira] [Created] (FLINK-10800) Abstract the StreamPartitionerTest for common codes

zhijiang created FLINK-10800: Summary: Abstract the StreamPartitionerTest for common codes Key: FLINK-10800 URL: https://issues.apache.org/jira/browse/FLINK-10800 Project: Flink Issue Type:

[jira] [Commented] (FLINK-10790) Refactor all the StreamPartitioner implementations into runtime module

[ https://issues.apache.org/jira/browse/FLINK-10790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16676092#comment-16676092 ] zhijiang commented on FLINK-10790: -- As I described above in this jira, the similar logics are

[jira] [Commented] (FLINK-10790) Refactor all the StreamPartitioner implementations into runtime module

[ https://issues.apache.org/jira/browse/FLINK-10790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16674875#comment-16674875 ] zhijiang commented on FLINK-10790: -- [~pnowojski], do you think it makes sense for unifying the

[jira] [Created] (FLINK-10790) Refactor all the StreamPartitioner implementations into runtime module

zhijiang created FLINK-10790: Summary: Refactor all the StreamPartitioner implementations into runtime module Key: FLINK-10790 URL: https://issues.apache.org/jira/browse/FLINK-10790 Project: Flink

[jira] [Updated] (FLINK-10537) Network throughput performance regression after broadcast changes

[ https://issues.apache.org/jira/browse/FLINK-10537?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10537: - Issue Type: Sub-task (was: Bug) Parent: FLINK-10745 > Network throughput performance

[jira] [Updated] (FLINK-9913) Improve output serialization only once in RecordWriter

[ https://issues.apache.org/jira/browse/FLINK-9913?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-9913: Issue Type: Sub-task (was: Improvement) Parent: FLINK-10745 > Improve output serialization only

[jira] [Updated] (FLINK-10662) Refactor the ChannelSelector interface for single selected channel

[ https://issues.apache.org/jira/browse/FLINK-10662?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10662: - Issue Type: Sub-task (was: Improvement) Parent: FLINK-10745 > Refactor the ChannelSelector

[jira] [Updated] (FLINK-10321) Make the condition of broadcast partitioner simple

[ https://issues.apache.org/jira/browse/FLINK-10321?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10321: - Issue Type: Sub-task (was: Improvement) Parent: FLINK-10745 > Make the condition of broadcast

[jira] [Created] (FLINK-10745) Improve record writer and serializer stack

zhijiang created FLINK-10745: Summary: Improve record writer and serializer stack Key: FLINK-10745 URL: https://issues.apache.org/jira/browse/FLINK-10745 Project: Flink Issue Type: Improvement

[jira] [Commented] (FLINK-10661) Initial credit should be configured in a separate parameter

[ https://issues.apache.org/jira/browse/FLINK-10661?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16671068#comment-16671068 ] zhijiang commented on FLINK-10661: -- [~NicoK], do you have any other suggestions for this issue? If not,

[jira] [Updated] (FLINK-10653) Fine-grained Shuffle System

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Description: This is the umbrella issue for improving shuffle for batch jobs. The shuffle behavior in

[jira] [Updated] (FLINK-10653) Fine-grained Shuffle System

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Summary: Fine-grained Shuffle System (was: Propose of external shuffle service for batch jobs) >

[jira] [Updated] (FLINK-10653) Propose of external shuffle service for batch jobs

[ https://issues.apache.org/jira/browse/FLINK-10653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] zhijiang updated FLINK-10653: - Description: This is the umbrella issue for improving batch shuffle.   The network shuffle behavior

<    5   6   7   8   9   10   11   12   >