ulysses-you commented on code in PR #6380:
URL: https://github.com/apache/incubator-gluten/pull/6380#discussion_r1671479487
##########
gluten-celeborn/common/src/main/java/org/apache/spark/shuffle/gluten/celeborn/CelebornShuffleManager.java:
##########
@@ -132,6 +142,11 @@ private SparkShuffleManager
vanillaCelebornShuffleManager() {
if (_vanillaCelebornShuffleManager == null) {
synchronized (this) {
if (_vanillaCelebornShuffleManager == null) {
+ if ("none"
+ .equalsIgnoreCase(
+ conf.get(SPARK_CELEBORN_COMPRESSION_CODEC_KEY,
celebornDefaultCodec))) {
+ conf.set(SPARK_CELEBORN_COMPRESSION_CODEC_KEY,
celebornDefaultCodec);
Review Comment:
Shall we use a new spark conf copy to hold the configs for row-based
celeborn shuffle manager ? I think it's risk to share the configs for both
row-based and columnar shuffle manager, for example, the first shuffle is
row-based then the conf will overwrite, and later the columnar shuffle manager
would use the overwritten conf.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
