pyspark - How costly is unpersist operation on spark RDDs? -
i wondering, how costly can berdd.unpersist()
operation on spark rdds ?, , storage level setting impact performance of operation ? benchmark(results/technique) full.
unpersist
releases rdd cache (memory , disk) , deletes shuffle files depends on. needs send message executors. should cheapest operation can rdd — not worth benchmarking.
note when rdd garbage collected, unpersist
automatically called on it. cannot avoid cost anyway.
Comments
Post a Comment