Hi everybody, since I am new to Spark, I am familiarizing with it by writing CPU-intensive applications like kmeans and knn. However, I observe some threads other than the worker threads using a lot of CPU. In particular, in jvisualvm, I observe the Acceptor and qtp threads to show such behavior, even during the computational phases. To be sure about the issue, I increased the computational time (by increasing the cached input size), which now takes more than 1 minute, and even throughout this longer map phase the CPU usage of these threads still shows up, although this phase does not require any communication meanwhile. Has anyone encountered this issue before? What might be the reason for it? How can I fix it? Thank you,
Alberto -- View this message in context: http://apache-spark-user-list.1001560.n3.nabble.com/high-CPU-usage-for-acceptor-and-qtp-threads-tp26053.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
