[GitHub] [flink] curcur commented on a change in pull request #13648: [FLINK-19632] Introduce a new ResultPartitionType for Approximate Local Recovery

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

curcur commented on a change in pull request #13648:
URL: https://github.com/apache/flink/pull/13648#discussion_r508193083



##########
File path: flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/ResultPartitionType.java
##########
@@ -71,7 +71,17 @@
 	 * <p>For batch jobs, it will be best to keep this unlimited ({@link #PIPELINED}) since there are
 	 * no checkpoint barriers.
 	 */
-	PIPELINED_BOUNDED(true, true, true, false);
+	PIPELINED_BOUNDED(true, true, true, false),
+
+	/**
+	 * Pipelined partitions with a bounded (local) buffer pool to support downstream task to
+	 * continue consuming data after reconnection in Approximate Local-Recovery.
+	 *
+	 * <p>Pipelined results can be consumed only once by a single consumer at one time.
+	 * {@link #PIPELINED_APPROXIMATE} is different from {@link #PIPELINED_BOUNDED} in that
+	 * {@link #PIPELINED_APPROXIMATE} is not decomposed automatically after consumption.
+	 */
+	PIPELINED_APPROXIMATE(true, true, true, true);

Review comment:
       It is similar to `bounded` in `Pipelined_Bounded`: use a fixed limit on the buffer pool size




----------------------------------------------------------------
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