[GitHub] [incubator-pinot] buchireddy commented on a change in pull request #4362: Optimize MutableOffHeapByteArrayStore by directly calling the PinotDataBuffer API

[GitBox <gi...@apache.org>]

buchireddy commented on a change in pull request #4362: Optimize MutableOffHeapByteArrayStore by directly calling the PinotDataBuffer API
URL: https://github.com/apache/incubator-pinot/pull/4362#discussion_r297502034
 
 

 ##########
 File path: pinot-core/src/main/java/org/apache/pinot/core/io/writer/impl/MutableOffHeapByteArrayStore.java
 ##########
 @@ -113,41 +106,41 @@ private int add(byte[] value) {
         // full
         return -1;
       }
-      for (int i = 0, j = startOffset; i < value.length; i++, j++) {
-        _byteBuffer.put(j, value[i]);
-      }
-      _byteBuffer.putInt(_numValues * Integer.BYTES, startOffset);
+      _pinotDataBuffer.readFrom(startOffset, value);
 
 Review comment:
   I'm also curious if this gives performance improvement and how much. As per the code in `org.apache.pinot.core.segment.memory.PinotByteBuffer#readFrom(long, byte[], int, int)`, if the buffer size is >10 (hard-coded value), the buffer is indeed duplicated during the batch operation so there could be garbage impact too.

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


With regards,
Apache Git Services

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