大数据集群间数据由hbase到impala

10.0.0.11
创建测试表
create 'LV_TEST',{NAME =>'cf' ,BLOOMFILTER => 'ROW',COMPRESSION => 'SNAPPY',VERSIONS => 1}
插入测试数据
put 'LV_TEST','rowkey001','cf:name','zhangsan'
put 'LV_TEST','rowkey001','cf:age','30'
get 'LV_TEST','rowkey001', {COLUMN=>'cf:name'}
get 'LV_TEST','rowkey001'
put 'LV_TEST','rowkey002','cf:name','lisi'
put 'LV_TEST','rowkey002','cf:age','40'
get 'LV_TEST','rowkey002'
查看测试表
scan 'LV_TEST'
导出到HDFS
find ./ /opt -name hbase
cd /opt/cloudera/parcels/CDH-5.5.1-1.cdh5.5.1.p0.11/bin
hbase org.apache.hadoop.hbase.mapreduce.Export LV_TEST /data/LV_TEST
hadoop fs -ls /data/LV_TEST
hadoop fs -get /data/LV_TEST
然后通过FTP下载到本地
上传到另一服务器
192.168.51.251
上传到HDFS
su - hdfs
hadoop fs -mkdir /data
hadoop fs -put /LV_TEST /data/LV_TEST
hadoop fs -ls  /data/LV_TEST
创建Hbase表
create 'LV_TEST_COPY',{NAME =>'cf' ,BLOOMFILTER => 'ROW',COMPRESSION => 'SNAPPY',VERSIONS => 1}
scan 'LV_TEST_COPY'
hbase org.apache.hadoop.hbase.mapreduce.Import LV_TEST_COPY /data/LV_TEST


建立hive外表
hive
CREATE EXTERNAL TABLE LV_TEST_COPY(id string,name string,age string)
ROW FORMAT SERDE 'org.apache.hadoop.hive.hbase.HBaseSerDe'
STORED BY 'org.apache.hadoop.hive.hbase.HBaseStorageHandler'
WITH SERDEPROPERTIES ("hbase.columns.mapping" = ":key, cf:name, cf:age")
TBLPROPERTIES("hbase.table.name" = "LV_TEST_COPY");


impala
impala-shell -i node90
同步元数据
INVALIDATE METADATA;
SQL查询结果

select * from LV_TEST_COPY;

最后将外部表数据复制到基于parques的impala内部表，完成数据转换。

insert into ***() select  * from *****