Hi Andrey,

Below is the description of MEMORY_ONLY fromĀ https://spark.apache.org/docs/latest/rdd-programming-guide.html

"Store RDD as deserialized Java objects in the JVM. If the RDD does not fit in memory, some partitions will not be cached and will be recomputed on the fly each time they're needed. This is the default level."

Just curious how do you know Spark will be disk even option MEMORY_ONLY is chosen?



On Sun, Mar 3, 2019 at 1:47 PM Andrey Dudin <dudin.andrey@gmail.com> wrote:
Hello everyone,

Is there a way to prevent caching data to disk even if the memory(RAM) runs out?
As I know, spark will use disk even if I use MEMORY_ONLY. How to disable this mechanism? I want to get something like out of memory exception if the memory(RAM) runs out.