HBase Data Model

Table & Column Accessibility
Row Keytimestampcolumn Accessibility
Urihttp://www.aliyun.com/zixun/aggregation/33959.html ">parser
R1t3url=http://www.taobao.comtitle= Daily Specials
T2host=taobao.com
T1
R2t5url=http://www.alibaba.comcontent= every day ...
T4host=alibaba.com
? Row key: Line keys, primary keys for table, records in table sorted by row key
? Timestamp: Timestamp, time stamp for each data operation, can be considered as the version number of the data
? Column accessibility: a clustered table that has one or more column accessibility in the horizontal direction, one column accessibility can consist of any number of columns, that is, column accessibility supports dynamic expansion. There is no need to define the number and type of column, all of the column is stored in binary format, and the user needs to do the type conversion.
Table & Region
When the table becomes larger as the number of records increases, it gradually splits into multiple splits, becoming regions, and a region is represented by [Startkey,endkey], Different region are assigned to the corresponding regionserver by master:

-root-&&. META. Table
HBase has two special table,-root-and. META.
? . Meta.: Region information for the user table is recorded. Meta. can have multiple Regoin
? -root-: recorded. META. Table region information,-root-only one region
? The location of the-root-table is recorded in the zookeeper

Client access to user data requires first access to zookeeper, then access to-root-tables, and then access. META. Table, the last to find the location of user data to access, the middle of a number of network operations, but the client side will do cache caching.
MapReduce on HBase
The most convenient and practical model for running batch operations on HBase systems remains mapreduce, as shown in the following illustration:

The relationship between HBase table and region is similar to that of HDFs file and block, HBase provides a matching Tableinputformat and Tableoutputformat API that can easily HBase Table as the source and sink of Hadoop mapreduce, for MapReduce job application developers, there is little need to focus on the details of the hbase system itself.