This chapter explored JSON support in SQL Server 2017. It is not as robust and deep as  XML—there is no native data type, and no optimized storage, and therefore you cannot create JSON indexes to improve performance. Thus, we are talking about built-in and not native JSON support.

However, even with built-in support, it is easy and handy to integrate JSON data in SQL Server. For most of JSON data processing, it would be acceptable. For large JSON documents stored in large database tables, it would be more appropriate to use DocumentDB or other NoSQL based solutions.

In this chapter, you learned that SQL Server 2017 brings built-in support for JSON data; unlike XML, there is no native data type. You used the FOR JSON extension to generate JSON from data in a tabular format and converted JSON data into a tabular format by using the OPENJSON rowset function. You learned how to parse, query, and modify JSON data with a function and how to improve the performance of JSON data processing by using indexes on computed columns and full-text indexes. You also discovered the limitations of JSON implementation in SQL Server 2017.

In the next chapter, you will meet another promising SQL Server feature—Stretch Databases.