spark优化

8 Performance Optimization Techniques Using Spark

https://www.syntelli.com/eight-performance-optimization-techniques-using-spark#

Spark性能优化指南(美团)

https://tech.meituan.com/2016/04/29/spark-tuning-basic.html

https://tech.meituan.com/2016/05/12/spark-tuning-pro.html

0 分析web ui

分析时间的消耗

1.多个map合并(??)

1
rdd1.map().map() ->  rdd1.map() 

2.减少action算子

说白了就是多个action操作,transformation逻辑可以写一起,最后action

3 增加分区 增加并行度

分数是说rdd分区

并行度是说executor num*executor core num

Author

Lavine Hu

Posted on

2022-01-05

Updated on

2024-04-05

Licensed under

Comments

:D 一言句子获取中...