[ 
https://issues.apache.org/jira/browse/FLINK-8360?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16374336#comment-16374336
 ] 

ASF GitHub Bot commented on FLINK-8360:
---------------------------------------

Github user tillrohrmann commented on a diff in the pull request:

    https://github.com/apache/flink/pull/5239#discussion_r170249880
  
    --- Diff: 
flink-runtime/src/main/java/org/apache/flink/runtime/state/TaskExecutorLocalStateStoresManager.java
 ---
    @@ -186,6 +210,34 @@ private void doRelease(Iterable<TaskLocalStateStore> 
toRelease) {
                return localRecoveryMode;
        }
     
    +   private File[] allocationBaseDirectories(AllocationID allocationID) {
    +           File[] allocationDirectories = new 
File[localStateRootDirectories.length];
    +           for (int i = 0; i < localStateRootDirectories.length; ++i) {
    +                   allocationDirectories[i] = new 
File(localStateRootDirectories[i], allocationSubDirString(allocationID));
    --- End diff --
    
    `allocationSubDirString` could be pulled out of the loop.


> Implement task-local state recovery
> -----------------------------------
>
>                 Key: FLINK-8360
>                 URL: https://issues.apache.org/jira/browse/FLINK-8360
>             Project: Flink
>          Issue Type: New Feature
>          Components: State Backends, Checkpointing
>            Reporter: Stefan Richter
>            Assignee: Stefan Richter
>            Priority: Major
>             Fix For: 1.5.0
>
>
> This issue tracks the development of recovery from task-local state. The main 
> idea is to have a secondary, local copy of the checkpointed state, while 
> there is still a primary copy in DFS that we report to the checkpoint 
> coordinator.
> Recovery can attempt to restore from the secondary local copy, if available, 
> to save network bandwidth. This requires that the assignment from tasks to 
> slots is as sticky is possible.
> For starters, we will implement this feature for all managed keyed states and 
> can easily enhance it to all other state types (e.g. operator state) later.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to