Handling Customized, JSON, and Blob Data Types
Some database management systems offer a way of storing niche column types such as JSON and Blob-related data. These column types are useful for rapid prototyping, handling schemaless data, and sending and receiving buffered data.
Typically, an application would use a NoSQL database, such as MongoDB, to process and query JSON documents, but this comes with a set of its own problems. We can no longer adhere to some sort of normalization for our structures without an extensive list of validations, and the NoSQL database cannot perform transactions nor provide ACID-compliant capabilities.
Note
Some NoSQL databases claim to offer ACID compliance, but they often come with stipulations and limitations such as a maximum number of documents that can be updated in a single transaction, or a transaction cannot take longer than some temporal window; otherwise, you will lose all of the performance advantages of NoSQL over SQL databases.
...