What is the best way to implement information with a one to many relationship in a large automation system located on a local server? Getting a report and record it is the most important point of the system.
For example, consider a travel agency that has several guides that perform one tour in week. To record the agency's information, best way is use three tables, one to store the agency's information and one to store personalized guide information and one to store relationship(one to many) but whit this way, changing the person's guide information that will perform a tour this week will change, so the report received from the system will also change. Another method of storing is to use a table, in which the agency's information is stored, and the person's information is also stored in a column in the json format.For example, consider the following table.
Agency Table: agency_id | name | phone_number | address | guides_details (That information stored in this column is stored in json format like name, mobile and ....)
Now my question is: Given the fact that the storage size for the project is not important (due to the use on the local server), is the second method suitable for large-scale information storage (archive)? Or, if there's another way to archive it, teach me please.