[ 
https://issues.apache.org/jira/browse/HADOOP-17461?focusedWorklogId=789476&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-789476
 ]

ASF GitHub Bot logged work on HADOOP-17461:
-------------------------------------------

                Author: ASF GitHub Bot
            Created on: 11/Jul/22 11:26
            Start Date: 11/Jul/22 11:26
    Worklog Time Spent: 10m 
      Work Description: steveloughran commented on code in PR #4352:
URL: https://github.com/apache/hadoop/pull/4352#discussion_r917816203


##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/statistics/impl/IOStatisticsContext.java:
##########
@@ -51,21 +51,21 @@ public class IOStatisticsContext {
   private static final boolean IS_THREAD_IOSTATS_ENABLED;
 
   private static final WeakReferenceThreadMap<IOStatisticsContext>

Review Comment:
   let's add a brief comment here as it is a key part of the design



##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/statistics/impl/IOStatisticsContext.java:
##########
@@ -164,13 +165,25 @@ public void setThreadIOStatistics(
    *
    * @return IOStatisticsSnapshot of the current thread.
    */
-  public IOStatisticsSnapshot getThreadIOStatisticsSnapshot() {
+  public IOStatisticsSnapshot snapshotCurrentThreadIOStatistics() {
     if (IS_THREAD_IOSTATS_ENABLED) {
       return (IOStatisticsSnapshot) getThreadIOStatistics();
     }
     return new IOStatisticsSnapshot();
   }
 
+  /**
+   * Reset the thread IOStatistics for current thread.
+   */
+  public void resetThreadIOStatisticsForCurrentThread() {
+    if (IS_THREAD_IOSTATS_ENABLED) {
+      IOStatisticsSnapshot currentThreadIOStatsSnapshot =
+          (IOStatisticsSnapshot) getThreadIOStatistics();
+      currentThreadIOStatsSnapshot.clear();
+      setThreadIOStatistics(currentThreadIOStatsSnapshot);

Review Comment:
   don't think this is needed



##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/statistics/impl/IOStatisticsContext.java:
##########
@@ -51,21 +51,21 @@ public class IOStatisticsContext {
   private static final boolean IS_THREAD_IOSTATS_ENABLED;
 
   private static final WeakReferenceThreadMap<IOStatisticsContext>
-      ACTIVE_IOSTATS_CONTEXT = new WeakReferenceThreadMap<>(
-      IOStatisticsContext::createNewInstance,
-      IOStatisticsContext::referenceLostContext
+      ACTIVE_IOSTATS_CONTEXT =
+      new WeakReferenceThreadMap<>(IOStatisticsContext::createNewInstance,
+          IOStatisticsContext::referenceLostContext
   );
 
   /**
    * Collecting IOStatistics per thread.
    */
   private final WeakReferenceThreadMap<IOStatisticsAggregator>

Review Comment:
   now we are casting to a snapshot, why not make this of type 
`IOStatisticsSnapshot`



##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/statistics/impl/IOStatisticsContext.java:
##########
@@ -164,13 +165,25 @@ public void setThreadIOStatistics(
    *
    * @return IOStatisticsSnapshot of the current thread.
    */
-  public IOStatisticsSnapshot getThreadIOStatisticsSnapshot() {
+  public IOStatisticsSnapshot snapshotCurrentThreadIOStatistics() {
     if (IS_THREAD_IOSTATS_ENABLED) {
       return (IOStatisticsSnapshot) getThreadIOStatistics();
     }
     return new IOStatisticsSnapshot();
   }
 
+  /**
+   * Reset the thread IOStatistics for current thread.
+   */
+  public void resetThreadIOStatisticsForCurrentThread() {
+    if (IS_THREAD_IOSTATS_ENABLED) {
+      IOStatisticsSnapshot currentThreadIOStatsSnapshot =
+          (IOStatisticsSnapshot) getThreadIOStatistics();

Review Comment:
   This is going to demand create a snapshot context even if there wasn't one 
needing resetting. How about you just look in the reference map (via 
lookup(currentThreadId())); and only if a ref is returned do the resetting?





Issue Time Tracking
-------------------

    Worklog Id:     (was: 789476)
    Time Spent: 2h 50m  (was: 2h 40m)

> Add thread-level IOStatistics Context
> -------------------------------------
>
>                 Key: HADOOP-17461
>                 URL: https://issues.apache.org/jira/browse/HADOOP-17461
>             Project: Hadoop Common
>          Issue Type: Sub-task
>          Components: fs, fs/azure, fs/s3
>    Affects Versions: 3.3.1
>            Reporter: Steve Loughran
>            Assignee: Mehakmeet Singh
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 2h 50m
>  Remaining Estimate: 0h
>
> For effective reporting of the iostatistics of individual worker threads, we 
> need a thread-level context which IO components update.
> * this contact needs to be passed in two background thread forming work on 
> behalf of a task.
> * IO Components (streams, iterators, filesystems) need to update this context 
> statistics as they perform work
> * Without double counting anything.
> I imagine a ThreadLocal IOStatisticContext which will be updated in the 
> FileSystem API Calls. This context MUST be passed into the background threads 
> used by a task, so that IO is correctly aggregated.
> I don't want streams, listIterators &c to do the updating as there is more 
> risk of double counting. However, we need to see their statistics if we want 
> to know things like "bytes discarded in backwards seeks". And I don't want to 
> be updating a shared context object on every read() call.
> If all we want is store IO (HEAD, GET, DELETE, list performance etc) then the 
> FS is sufficient. 
> If we do want the stream-specific detail, then I propose
> * caching the context in the constructor
> * updating it only in close() or unbuffer() (as we do from S3AInputStream to 
> S3AInstrumenation)
> * excluding those we know the FS already collects.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to