klsince commented on code in PR #11020:
URL: https://github.com/apache/pinot/pull/11020#discussion_r1253587772


##########
pinot-segment-local/src/main/java/org/apache/pinot/segment/local/upsert/BaseTableUpsertMetadataManager.java:
##########
@@ -75,6 +115,145 @@ public void init(TableConfig tableConfig, Schema schema, 
TableDataManager tableD
 
     _enableSnapshot = upsertConfig.isEnableSnapshot();
     _serverMetrics = serverMetrics;
+    _helixManager = helixManager;
+    _preloadExecutor = preloadExecutor;
+    try {
+      if (_enableSnapshot && upsertConfig.isEnablePreload()) {
+        preloadSegments();

Review Comment:
   In fact, the `createTableDataManager()` method pasted above creates the 
TableDataManager object first, and then calls its init() method. The preloading 
logic happens as part of the init() method. So among the Helix threads about to 
handle segment state transitions, only the first Helix thread would proceed and 
preload segments, and the other Helix threads would wait for it to finish, as 
ensured by ConcurrentHashMap.computeIfAbsent(... lambda), which runs the lambda 
atomically



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org
For additional commands, e-mail: commits-h...@pinot.apache.org

Reply via email to