sry, my meaning the table in HBase is below , please ignore the one in last mail.
*row*** *Column families* *Info:*** *student_name:*** *student_type:* <course_id> Info:name student_name:<student_id> student_type:<student_id> Info:gender Info:age ---------- Forwarded message ---------- From: SingoWong <singoch...@gmail.com> Date: Fri, May 14, 2010 at 4:26 PM Subject: About HBase schema design To: hbase-user@hadoop.apache.org Hi, I begin HBase schema design, but now i face some confused. In RDBMS, my design is 3 tables: * Courses id pk name type Students id pk curses_id fk => Courses.id name gender age * * * In HBase, i think my design is *row*** *Column families* *Info:*** *Course_name:*** *Course_type:* <student_id> Info:name Course_name:<course_id> Course_type:<course_id> Info:gender Info:age Is it my design correct? but if design like this, when i want to find how many students under one course, i dont know how to filter. if i design as two table in hbase, is it normalization? Thanks & Regards, Singo
