ksmou commented on code in PR #9925:
URL: https://github.com/apache/hudi/pull/9925#discussion_r1374005472
##########
hudi-client/hudi-client-common/src/main/java/org/apache/hudi/config/HoodieClusteringConfig.java:
##########
@@ -161,6 +161,13 @@ public class HoodieClusteringConfig extends HoodieConfig {
+ "value will let the clustering job run faster, while it will give
additional pressure to the "
+ "execution engines to manage more concurrent running jobs.");
+ public static final ConfigProperty<Integer>
CLUSTERING_READ_RECORDS_PARALLELISM = ConfigProperty
+ .key("hoodie.clustering.read.records.parallelism")
+ .defaultValue(20)
Review Comment:
We verify that 20 is a relatively stable parameter running job. The param
hoodie.clustering.plan.strategy.target.file.max.bytes is 1g which contains many
files, we read it with 20 parallelism is enough and no need to increase
executor-memory.
We should limit the parallelism by default, this can reduce the number of
tasks to a great extent.
Compaction has a similiar use, it parallelize operations with its size.
```
// HoodieCompactor.java
return context.parallelize(operations).map(operation -> compact(
compactionHandler, metaClient, config, operation,
compactionInstantTime, maxInstantTime, instantRange, taskContextSupplier,
executionHelper))
.flatMap(List::iterator);
```
--
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]
