Is the number of indexes in a table the more the better?

Go to InnoDB When adding data to the table , The cluster index will be automatically established based on the primary key . As we keep inserting data into the table , Will keep inserting data into the data page . When a data page is full , It will split into multiple data pages , At this time, the index page is needed to point to each data page .

If there are too many data pages , Then there will be too many data page pointers in the index page , The index page is bound to be full , So the index page will split , Then form a higher index page .

Default MySQL The cluster indexes established are organized based on the value of the primary key , The leaf nodes of cluster index are all data pages , Inside is the inserted row of complete data .

• Data pages / The records in the index page form a one-way linked list , Ordered by data size
• Data pages / Index pages form a two-way linked list with each other , They are also arranged in order according to the size of the data

therefore B+ An index tree is a completely ordered data structure , Whether inside or between pages , This allows you to find data , Start directly from the root node and look down layer by layer according to the size of the data value , It's very efficient .

MySQL Create indexes corresponding to some fields in the table of , What's good about it ？

It can be directly based on the index of a field B+ Tree to find data , No need for full table search , High performance improvement . But there are downsides ：

Space

Every B+ Trees take up a lot of disk space , Too many indexes , It takes a lot of disk space .

Do a lot of indexing , When adding, deleting, modifying and checking , Maintain the data order of each index every time , Because every index B+ Trees require that pages be sorted by value size , Pages are also ordered ：

$NextOneindividualpageOftheYesvaluehave\; toMust\; be＞OnOneindividualpageOftheYesvalue$

So keep adding, deleting, modifying and checking , It will inevitably lead to the value size between various data pages may be out of order , For example, a smaller value is inserted in the next data page , It's smaller than the value of the previous data page ！ At this time, there is no way , Only data pages can be moved , Maintain the order between pages .

Time

Keep inserting data , The data pages of each index keep splitting , Keep adding new index pages , This process is time-consuming .

So if you have too many indexes in a table , It is likely to lead to the poor speed of your addition, deletion and modification , Maybe the query speed can be improved , But the addition, deletion and modification will be affected , Therefore, it is not recommended to have too many indexes in a table ！