This might sound like a naive question but when are database table indexes in mysql required? How do such indexes affect performance and why?
- Anybody can ask a question
- Anybody can answer
- The best answers are voted up and rise to the top
closed as off topic by Jim G., MainMa, Yusubov, Glenn Nelson, Walter Dec 23 '12 at 16:24
Questions on Programmers Stack Exchange are expected to relate to software development within the scope defined by the community. Consider editing the question or leaving comments for improvement if you believe the question can be reworded to fit within the scope. Read more about reopening questions here.If this question can be reworded to fit the rules in the help center, please edit the question.
The purpose of database indexes is to increase the performance of searches in data tables.
As you might already know, for unsorted data structures, the asymptotic notation O(f(n)) for searches is O(n). Meaning that, in the worst case, all rows of the table are going to be looked in order to rows matching the WHERE clause (and the same goes for JOIN columns). It might not be that bad in a small table, but it can become really slow on a bigger table.
Since a data table cannot physically contain sorted data for multiple columns (generally speaking, it may only be physically sorted by a single column), indexes become handy. In summary, indexes are usually a B-tree (or a variant) sorted map containing a specific column data as key and the row information as the entry value. This way, the O(f(n)) can be reduced to O(log(n)).
Are indexes required? No. But your queries will show a huge performance boost, if you use indexes in the right way.
Yet the performance gain has its drawbacks: in order to increase the search performance, the insert and delete operations become slower (as the indexes must be updated). Here are some asymptotic amortized worst-case analysis for different data structures. Remember this is just an introduction on the matter. You should definitely take a deeper look on database indexes.
Basically, an index on a column allows you to very quickly find all rows of a table that have a specific value (or range of values) in that column, without having to look at every single row, and no matter how large the table is.
Thus, the index can drastically speed up queries which have a WHERE clause referring to that column, as well as JOINs and nested SELECTs. Especially for very large tables, indexes are absolutely essential to get good performance. But of course, they only help if you have them on the columns your queries actually use.