【大数据】hadoop计数器

org.apache.hadoop.mapreduce.FileSystemCounter

FILE_BYTES_READ

FILE_BYTES_WRITTEN

FILE_READ_OPS

FILE_LARGE_READ_OPS

FILE_WRITE_OPS

HDFS_BYTES_READ

HDFS_BYTES_WRITTEN

HDFS_READ_OPS

HDFS_LARGE_READ_OPS

HDFS_WRITE_OPS                  

org.apache.hadoop.mapreduce.JobCounter

TOTAL_LAUNCHED_MAPS

TOTAL_LAUNCHED_REDUCES

RACK_LOCAL_MAPS

SLOTS_MILLIS_MAPS

SLOTS_MILLIS_REDUCES

FALLOW_SLOTS_MILLIS_MAPS

FALLOW_SLOTS_MILLIS_REDUCES  

org.apache.hadoop.mapreduce.TaskCounter

MAP_INPUT_RECORDS

MAP_OUTPUT_RECORDS

MAP_OUTPUT_BYTES

SPLIT_RAW_BYTES

COMBINE_INPUT_RECORDS

COMBINE_OUTPUT_RECORDS

REDUCE_INPUT_GROUPS

REDUCE_SHUFFLE_BYTES

REDUCE_INPUT_RECORDS

REDUCE_OUTPUT_RECORDS

SPILLED_RECORDS

CPU_MILLISECONDS

PHYSICAL_MEMORY_BYTES

VIRTUAL_MEMORY_BYTES

COMMITTED_HEAP_BYTES  

org.apache.hadoop.mapreduce.lib.input.FileInputFormatCounter

BYTES_READ

举例：

jobClient.getJob(jobName).getCounters().getGroup("org.apache.hadoop.mapreduce.FileSystemCounter").getCounter("HDFS_BYTES_WRITTEN")

本文转自 wws5201985 51CTO博客，原文链接：http://blog.51cto.com/yjplxq/1330128，如需转载请自行联系原作者