New answers tagged database-recommendation
Try these design which will satisfy your requirements. Group(Group Id Primary Key,Group Name) User(User Id Primary Key,Name,Email,Created on,Group Id references Group(Group Id),Referred User ID references User(User ID)) Task(Task ID Primary Key,Name,Priority,Date) User-Tasks(Relation ID Primary Key,User ID references User(User ID),Task ID references ...
Start understanding that the problems you face have nothing to do with the database technology you use. They are caused by physics, and physics does not care about oracle, microsoft or open source. It is the same. Generally - depending on the queries you may want a HUGH server (albeit "billions of rows" sounds too much for a phone dictionary). Or cluster of ...
You could do that. I'm thinking something along the lines of having an active and inactive date on your fact table for each product for each customer. Then you could mark a previous product with an inactive date. This should allow you to query it for date ranges that are include a specific date to get what their product list was for a specific point in ...
If you have multiple product table for each customer then your you should write an Indexed View to get these figures instead of going into Fact Table. Even if you have a single table for products for customers, then you should go for an indexed view.
I think you are looking for DBs that are designed for analytics. May be you should check these solutions:. some data storage solutions specific to realtime analytics Druid Spark Cassandra Hbase hypertable accumulo reference: http://kkovacs.eu/cassandra-vs-mongodb-vs-couchdb-vs-redis http://relistan.com/cassandra-vs-mongo/
While I work for a competitor of MongoDB, FatDB, but given your requirements are Java, I would say: Cassandra Google BigTable (not sure of Node.js support) Cloudera (more of a Hadoop platform)
Top 50 recent answers are included