From time to time we see messages as jvm pause defected from HiveMetastore & Hiveserver2 logs in spite we increased the heap size to the right size ( according to number of connection to HIVE reference - https://www.linkedin.com/pulse/tuning-apache-hive-cdh-sadique-arrvi/)
we don't sure if the default Garbage Collectors - Parallel , is good enough , and since our HIVE service is under huge load
so this make me to think about other Garbage Collectors as CMS or G1 ?
default Garbage Collector for HIVE - https://gennadny.wordpress.com/2020/11/02/dancing-with-the-elephants-and-flying-with-the-bees-apache-hive-performance-tuning/
Example from log's
2024-06-03 16:42:06,373 INFO [org.apache.hadoop.hive.common.JvmPauseMonitor$Monitor@40faff12]: common.JvmPauseMonitor (JvmPauseMonitor.java:run(193)) - Detected pause in JVM or host machine (eg GC): pause of approximately 2854ms
