Step 1: HDFS file data
[root@localhost mapr]# hadoop fs -cat //user/mapr/sample.txt
100,sachin,Mum,India,Cricket
200,Samantha,Chennai,india,Actress
300,Modi,Ahd,Ind,politician
Step 2: Creating table in HBase with rowkey
hbase(main):006:0> create 'blkimport','cf'
0 row(s) in 1.2990 seconds
=> Hbase::Table - blkimport
hbase(main):007:0> list
blkimport
hbase(main):010:0> scan 'blkimport'
ROW COLUMN+CELL
0 row(s) in 0.0180 seconds
Step 3: importing data from hdfs to HBase using ImportTsv
[mapr@localhost hbase-1.1.1]$ /opt/mapr/hbase/hbase-1.1.1/bin/hbase org.apache.hadoop.hbase.mapreduce.ImportTsv -Dimporttsv.separator=',' -Dimporttsv.columns='HBASE_ROW_KEY,cf:pname,cf:loc,cf:country,cf:prefession' blkimport /user/mapr/sample.txt
Here /user/mapr/sample.txt--> HDFS file
blkimport --------------> HBase table
cf------------------------> ColumnFamily
HBASE_ROW_KEY-> RowKey
pname,loc,country,prefession--> Column Names
Step 4: Scan table HBase
hbase(main):011:0> scan 'blkimport'
ROW COLUMN+CELL
100 column=cf:country, timestamp=1477509673574, value=India
100 column=cf:loc, timestamp=1477509673574, value=Mum
100 column=cf:pname, timestamp=1477509673574, value=sachin
100 column=cf:prefession, timestamp=1477509673574, value=Cricket
200 column=cf:country, timestamp=1477509673574, value=india
200 column=cf:loc, timestamp=1477509673574, value=Chennai
200 column=cf:pname, timestamp=1477509673574, value=Samantha
200 column=cf:prefession, timestamp=1477509673574, value=Actress
300 column=cf:country, timestamp=1477509673574, value=Ind
300 column=cf:loc, timestamp=1477509673574, value=Ahd
300 column=cf:pname, timestamp=1477509673574, value=Modi
300 column=cf:prefession, timestamp=1477509673574, value=politician
3 row(s) in 0.0570 seconds
Nice post,Thank you for info..
ReplyDeleteKeep sharing more posts.
big data online course