hadoop的mapReduce和Spark的shuffle过程的详解与对比及优化
https://blog.cloudera.com/blog/2015/01/improving-sort-performance-in-apache-spark-its-a-double/
https://issues.apache.org/jira/browse/SPARK-751
Sort-BasedShuffleinSpark: https://issues.apache.org/jira/browse/SPARK-2045
https://issues.apache.org/jira/browse/SPARK-2926
https://jaceklaskowski.gitbooks.io/mastering-apache-spark/content/spark-shuffle-ShuffleManager.html