You can scale Open Source databases (SQL and NoSQL) as much as a commercial version.
An example of this is MariaDB , which replaces MySQL for applications with high scalability and performance.
In addition, giant companies use NoSQL banks or even clustered SQL to provide the required data.
What usually differentiates a business solution is the amount of resources offered, which are vastly larger in banks like Oracle and SQL Server than in MySQL or PostgreSQL, for example.
In addition, companies like Microsoft and Oracle offer a stack of integrated solutions, while in Open Source solutions you would have to configure and integrate everything "manually", generating sometimes larger costs with lower benefits .
Remember, business is all about value for money. And it also involves the issue of having a "security" that the product will work. Then we came to the issue of technical support.
For a company, a free system that does not work is more expensive than a paid system that works. When someone pays an Oracle or SQL Server license he is buying the guarantee from these large companies that bugs will be promptly fixed and their inquiries will be resolved quickly,
Note, however, that this is common for "normal" development companies. IT giants like Google and others, who often create and support new technologies, can afford to use only Open Source because they have the workforce and resources needed to build their own infrastructure and endure all this.
At this point, it is advantageous for them to have a super specialized engineer to handle the problems of tools within the company itself. For most businesses, however, this is not an option.