What is the default input split size in Mapreduce Job ?

The default input split size in Mapreduce job is equal to the size of the HDFS block which is 256 MB as of hadoop-2.4.0 release. Each of these input splits are processed by a separate map task.