Cassandra架构和RDBMS架构之间的区别

Question

I am a Cassandra newbie, so please bear with me if its a trivial question. I have been studying about Cassandra from this book http://shop.oreilly.com/product/0636920010852.do?cmp=il-orm-ans-learnmore-0636920010852 Although it seems a pretty old book I got a rough idea of Cassandra's schema-less data model. However in relatively newer documentation I found about the schema-optional nature of Cassandra. As far as I understand Cassandra's main strength lies in its horizontal scalabilty and distributed storage. Its advantage over MySQL and other related RDBMS is it frees us from the problems posed by hundreds of outer joins in larger data. My question is, when we choose to create a schema and use the normal tables, rows and column structure and query in CQL how does it differ from MySQL? What are the basic differences betwwen a MySQL schema and Cassandra Schema. According to datastax documentation

For example, CQL is adapted to the Cassandra data model and architecture so there is still no allowance for SQL-like operations such as JOINs or range queries over rows on clusters that use the random partitioner. This reference describes CQL 2.0.0.

So in cases where we use a schema like approach in Cassandra how do we handle the necessity of JOINS? As I have never worked with Cassandra before I am not aware of how to handle the data in a distributed way.

Answer 1

That's a really old book, cassandra has changed a lot. I suggest you to take this course for free http://datastaxacademy.elogiclearning.com Joins are replaced by denormalization which is explained in the course.

There are many differences between cassandra and mysql, in mysql if your database gets bigger than a server you have to use other tools to manage multiple servers which is not an easy task to do. In Cassandra, database is distributed from the first place and your tables can get big enough to live on 1000s of servers. In Cassandra, you can add new columns and different rows can have different columns.