How we can do CRUD operations on complex data models in Cassandra?

Question

How we can do CRUD operations on complex data models in Cassandra?

• I have a project using NOSQL.
• I have a column family for my customers.
• The column family has just "id" at first.
• Then it will be updated by altering new columns.
• Count and type of columns for each customer could be different.
• Also, each column can include sub columns with ids again and it would be altered, too. So, they should be indexed. And documents are not useful for this issue.

I've read about NOSQL, and I've decided to use Cassandra. I will be thankful if you would answer this questions:

1. Is the above that possible?

2. How we can create and use CRUD operations on this column family?

3. If the answer of last question is true, what is the type of result of a query?

4. It will return some rows for each primary key (id)?

5. How we can manage that, to access a table like with no redundancy? because I don't now this summarizing should be manage in DBside or in code side.

Thank you for your help.