My jobs frequently run out of memory if the #of cores on an executor is too high, because each core launches a new parquet decompressor thread, which allocates memory off heap to decompress. Consequently, even with say 12 cores on an executor, depending on the memory, I can only use 2-3 to avoid OOMs when reading parquet files.
Ideally I would want to use all 12 cores, but limit the # of parquet decompresses to 2-3 per executor. Is there some way to do this? Thanks, Ankit -- View this message in context: http://apache-spark-user-list.1001560.n3.nabble.com/Limit-of-parallel-parquet-decompresses-tp22022.html Sent from the Apache Spark User List mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
