找到你要的答案

Setting spark memory allocations for extracting 125 Gb of data…ExecutorLostFailure
提取125 GB的数据…executorlostfailure设置火花内存分配

apache-spark  apache-spark-sql  hawq 

Why HDFS not preferred with applications that require low latency?
为什么不喜欢HDFS需要低延迟的应用?

hadoop  apache-spark  hdfs  hawq