我正在接近Hbase,我需要在同一张表中加载各种时间序列CSV。
我特别有:
first.csv
+---+---+----------+-------------------+
|_c0| ID| log| datetime|
+---+---+----------+-------------------+
| 0| 9| 4r8|2001-12-10 01:00:00|
| 1| 45| 223|2001-12-10 01:00:00|
| 2| 9| iu8|2002-11-01 03:00:00|
秒
+---+---+----------+-------------------+
|_c0| ID| message| datetime|
+---+---+----------+-------------------+
| 0| 9| ERROR|2001-12-10 01:00:00|
| 1| 45| SUCCESS|2001-12-10 01:00:00|
| 2| 9| SUCCESS|2002-11-01 03:00:00|
我要在具有以下SuperColumnFamily模式的Hbase中加载:
ROW_KEY | ID = 9 | ID = 45
+-------------------+---------------+---------------+
| log | message | log | message |
|2001-12-10 01:00:00+-----+---------+---------------+
| 4r8 | ERROR | 223 | SUCCESS |
+-------------------+---------------+
| log | message |
|2002-12-01 03:00:00+-----+---------+
| iu8 | SUCCESS |
+-------------------+---------------+
该怎么办? (MapR,Spark,CompleteBulkLoad?)