join - Help me to get better understanding of Digg's Cassandra data model -
http://about.digg.com/blog/looking-future-cassandra
i've found article digg's move cassandra. didn't author's idea of bucket pair (user,item). little more details on idea helpful me understand solution better.
thanks
it sounds using 1 row in super column family per user 1 super column per item; subcolumn item super column represents friend dugg item. @ least in pycassa, makes insert simple as:
column_family.insert(user, {item: {friend: ''}})
they have done couple of other ways, , i'm not sure chose.
one use standard column family, use (user,item) combination row key, , use 1 column per friend dugg item:
column_family.insert(user + item, {friend: ''})
another use standard column family, use (user) row key, , use (item, friend) combination column name:
column_family.insert(user, {item + friend: ''})
doesn't sound used, it's acceptable option well.
Comments
Post a Comment