Spark analysis the Log Log file, the load to the HDFS and two methods of loading a local file, what's the difference?
Is loaded into the analysis speed in the HDFS? But the load of HDFS is slow,

CodePudding user response:

The top, to give points,,,

CodePudding user response:

Loaded into the HDFS slower than the load local files, because to go online, HDFS for large data files for storage

CodePudding user response:

A block from HDFS load, it is to use partition parallel computing, or through better acceleration load data file format