spark通过降低cores数量来提高成功率,错误记录之memory limit

错误复现

当数据量达到几百G约一亿条进行计算时,在原本的资源分配条件下失败的几率很高,失败的错误基本如下

Container killed by YARN for exceeding memory limits. 16.9 GB of 16 GB physical memory
used. Consider boosting spark.yarn.executor.memoryOverhead

通常来讲,我们设计资源分配时一般指调整这几个参数:

set spark.executor.instances=8;
set spark.executor.cores=8;
set spark.executor

你可能感兴趣的:(Spark基础与提高,Hive,spark,hive,内存优化,physical,memory)