I'm working with many datasets, for example i have a main table with id, and 30 more tables They are relating to main with id, but I have just given me that I have columns specific for each type of event in main table, then for example the main table
id event_type data1 data2 data3 data4 data5 …….. data15
1 primary x x null null x null
2 primary x x null null x null
3 secundary x null x x null null
4 secundary x null x x null null
5 primary-1 x x x x x null
Then for example, data2 is exclusive to secundary, data3 no have info in primary...etc.
so my question is, what would be the best way of working, create a table, for each type of event, and then link it to event_type, or put all the columns in a single primary table,
** note
if i do the first option, i need to work with leftjoin, becouse i have other 30 tables linked to id, and i working with leftjoin, and anyway there will be null for any columns.
my question is for the best optimization, I want faster processing in less time