hadoop - Hbase quickly count number of rows -
right implement row count on resultscanner this
for (result rs = scanner.next(); rs != null; rs = scanner.next()) { number++; } if data reaching millions time computing large.i want compute in real time don't want use mapreduce
how count number of rows.
use rowcounter in hbase rowcounter mapreduce job count rows of table. utility use sanity check ensure hbase can read blocks of table if there concerns of metadata inconsistency. run mapreduce in single process run faster if have mapreduce cluster in place exploit.
$ hbase org.apache.hadoop.hbase.mapreduce.rowcounter <tablename> usage: rowcounter [options] <tablename> [ --starttime=[start] --endtime=[end] [--range=[startkey],[endkey]] [<column1> <column2>...] ]
Comments
Post a Comment