Alowator commented on code in PR #13409:
URL: https://github.com/apache/hudi/pull/13409#discussion_r2137104958
##########
hudi-flink-datasource/hudi-flink/src/main/java/org/apache/hudi/configuration/FlinkOptions.java:
##########
@@ -586,6 +586,35 @@ private FlinkOptions() {
.withDescription("Maximum memory in MB for a write task, when the
threshold hits,\n"
+ "it flushes the max size data bucket to avoid OOM, default 1GB");
+ @AdvancedConfig
+ public static final ConfigOption<Boolean> WRITE_BUFFER_SORT_ENABLED =
ConfigOptions
+ .key("write.buffer.sort.enabled")
+ .booleanType()
+ .defaultValue(false) // default no limit
+ .withDescription("Whether to enable buffer sort within append write
function.");
+
+ @AdvancedConfig
+ public static final ConfigOption<String> WRITE_BUFFER_SORT_KEYS =
ConfigOptions
+ .key("write.buffer.sort.keys")
+ .stringType()
+ .noDefaultValue() // default no limit
+ .withDescription("Sort keys concatenated by comma for append write
function.");
+
+ @AdvancedConfig
+ public static final ConfigOption<Long> WRITE_BUFFER_SIZE = ConfigOptions
+ .key("write.buffer.size")
+ .longType()
+ .defaultValue(1000L) // default no limit
+ .withDescription("Buffer size of each partition key for append write
function.");
+
+ @AdvancedConfig
+ public static final ConfigOption<Long> WRITE_TIMER_INTERVAL = ConfigOptions
Review Comment:
+1 to this idea, from my experience BinaryInMemorySortBuffer works fast,
cause it uses index to sort, also it reduces GC pressure. Store elements as
List is a bad idea
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
