zjf2012 edited a comment on issue #23560: [SPARK-26632][Core] Separate Thread Configurations of Driver and Executor URL: https://github.com/apache/spark/pull/23560#issuecomment-471834412 @attilapiros , We tuned a lot to get optimized result with 8 threads on old code. For your case, I just ran the SparkBenchmarkHPC again with 256 nodes and got below result. The overall time is 235 compared to 196 with 8 threads. { "experiment" : { "id" : "simplemap-spark-scala" }, "args" : { "src" : "", "dst" : "", "cores" : "256", "generate" : "true", "lazy" : "false", "blocks" : "131072", "block_size" : "1", "multiplier" : "1048576", "block_size_unit" : "MB", "nparts" : "1", "size" : "1", "nodes" : "256", "json" : "nlog256-131072-256m_16.json", "xml" : "", "functional" : "false", "hybrid" : "false", "imperative" : "false" }, "performance" : { "map" : "45.63785444", "shift" : "77.308619541", "reduce" : "42.042015014", "average" : "70.184881466", "overall" : "235.173370461" } }
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
