I have seen many solutions developed with multiple Tables inside the same file, which is very practical, especially if the number of Tables is limited.
On the other hand, I have seen some very sophisticated solutions made of 3-5 files linked together.
Probably because I started using FM when it was 4, using separated files is still a component of my instinct.
Updates are easier and faster. And in case of corrupted data, recovery is faster and easier since in most cases do not affect the all solution but only a specific file.
AT the end, the real matter is performance/reliability, especially with > 500k records.
Has anyone tested performance comparing one-file-multiple-tables vs. separated files ?
By instinct I would say it should be faster with one-file-multiple-tables, but is there a “significant” difference ?
The reason of my question is because, if I have a solution with some 50 Tables, the relationship graph is terrible so I tend to split the solution into 3-4 files, grouping Tables accordingly with their logical use/relationships.