I am migrating RDBMS tables to cassandra. We track customers who are subscribing to different categories. There are some categories already but some categories maybe added new incourse of time. Right now we are joining the tables.
details table1 - columns
custid, name , address, phone
details table2 - custid, cat1, cat2, cat3, cat4
category details - catid, catname, catregion, catdescription, iscatmanadatory
In Cassandra I am trying to keep the customer id and name as primary keys. I am planning to keep categories subscribed by customer in a map. But if any new categories are added, will collection columns create any bottlenecks ?
