Cassandra table structure suggestion and way of query -
i trying create following hierarchy: userid rowkey, hourly time series columns , inside each hourly column want have user specific information such hourly activity.
{ userid:long { timestamp:datetime{ pageview: integer, clicks:integer } }
i've read possible achieve using supercolumns
@ same time mentioned supercolumns
outdated right now. if true, alternatives can use?
could please provide me cql / java thrift example how should create , insert such type of structure in cassandra?
thanks!
you can user composite primary key this, add table creation cql query table. , can use counter column clicks.
create table user_click_by_hour( userid long, time_stamp timestamp, clicks int, pageview int, primary key(userid,time_stamp)
)
Comments
Post a Comment