I read the sentence (and partitioning by default) in the book "Designing Data-intensive Application" and I don't understand it clearly could anyone who has a big experience in databases obvious it for me.
The sentence:
In the late 2000s, nonrelational (NoSQL) databases started gaining popularity. They
aimed to improve upon the relational status quo by offering a choice of new data
models (see Chapter 2), and by including replication (Chapter 5) and partitioning (Chapter 6) by default
including replication (Chapter 5) and partitioning (Chapter 6) by default
Ok, I see how this sentence can be difficult to interpret.
When considering databases in terms of (Dr. Brewer and Dr. Fox's) CAP Theorem it's important to remember that RDBMSs are considered to be "CA" databases. That is, in the rule of "pick two" they support consistency and availability, at the cost of having to confine the data to a single instance (server, container, etc).
NoSQL ("Not only SQL) DBs are (usually) either "AP" or "CP," making specific design choices to be able to survive network or other hardware failures.
NoSQL Database make a partitioning by default when create a database?
So the better way to describe this, is to say that NoSQL DBs give you the tools to deploy in such a way as to be able to survive failures of one or more instances. They make it easy(er) to build a resilient data platform with multiple, distributed replicas (no single point of failure).
However, it comes down to implementation. Someone could still deploy a NoSQL database in a way which is not partition tolerant. For instance, if they were to stand-up Apache Cassandra or MongoDB on a single instance (server, container, etc), that would in fact not be a partition tolerant system. Thus, they would not be getting "partitioning by default."