dongjoon-hyun commented on a change in pull request #24577: [Spark-27664][CORE]
file status cache perf issue/branch 2.4
URL: https://github.com/apache/spark/pull/24577#discussion_r282971585
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala
##########
@@ -542,6 +542,17 @@ object SQLConf {
.longConf
.createWithDefault(250 * 1024 * 1024)
+ val HIVE_FILESOURCE_FILE_CACHE_CONCURRENCY_LEVEL =
+ buildConf("spark.sql.hive.filesourceFileCacheConcurrencyLevel")
+ .doc("Provide concurrency level for underlying guava cache. The default
is 1." +
+ " Higher value may give better performance, when the number of " +
+ "entries in the cache is very large. This imposes a limit on the size
of single entry" +
+ "in the cache, i.e. sizeLimit < totalCacheCapacity/concurrencyLevel.
See SPARK-27664.")
+ .intConf
+ .checkValue(value => value >= 1 && value <= 64, "The concurrency level
should be " +
+ "between 1 and 64.")
+ .createWithDefault(1)
Review comment:
Hi, @ScrapCodes . Could you propose this new feature to `master` branch
first?
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
