"And now, when I create an index on column_value. Will the column_value still be stored with the column_key or will Cassandra create an extra column?"
Cassandra will create a column family whose name is "index_name" to index the column_value for you. You can't have access to this column family though... On Wed, Apr 23, 2014 at 5:02 PM, Sebastian Schmidt <isib...@gmail.com>wrote: > Yes I mean a secondary index. > > For example: > > CREATE INDEX index_name ON table_name(column_value); > > Am 23.04.2014 17:01, schrieb DuyHai Doan: > > What do you mean by "index on column_value" ? Do you mean secondary index ? > > > On Wed, Apr 23, 2014 at 4:52 PM, Sebastian Schmidt <isib...@gmail.com>wrote: > >> And now, when I create an index on column_value. Will the column_value >> still be stored with the column_key or will Cassandra create an extra >> column? >> >> Am 23.04.2014 16:47, schrieb DuyHai Doan: >> >> The schema you just showed allows, for one row key (partition key), to >> have several distinct pairs of column_key/column_value. And that's exactly >> what you want ... >> >> >> >> >> On Wed, Apr 23, 2014 at 4:44 PM, Sebastian Schmidt <isib...@gmail.com>wrote: >> >>> Hi, >>> >>> I want to create a storage layout with CQL3 like this: >>> >>> row_key should be my row key >>> column_key should by my column key >>> column_value should be the value saved for the column key >>> >>> How can I achieve this? >>> >>> I figured that doing this: >>> >>> CREATE TABLE table_name (row_key BLOB, column_key BLOB, column_value >>> BLOB, PRIMARY KEY (row_key, column_key)); >>> >>> would make column value unique for each row key. >>> >>> But I want to store multiple key/value pairs per row. How can I do that >>> with CQL3? >>> >>> Kind regards, >>> Sebastian >>> >> >> >> > >
