At what volumes of the database does it make sense to switch to search engines like Lucene / Elastic?

Question

MySQL

How to understand the moment - are you crookedly using MySQL and OS or is it just too much data?
How much should the base be - 10Mb, 1Gb or 100Gb?
What volumes do you use Lucene/Elastic on?

Answer 1

on any where you need a quick full-text search with options

Answer 2

There is no such fixed size, it's more a matter of functionality. After all, the built-in searches in MySQL are rather weak and come down for the most part to a simple complete enumeration of all values ​​​​and the execution of regular expressions on them.
We, for example, use Sphinx all the time, even for the smallest projects. This engine allows you to combine search queries (which run on an adult index and are very fast) with other MySQL queries.

Answer 3

It is very important to understand the differences between mysql and elasticsearch here. 2nd is the search engine first. We use it in conjunction with mysql, because transactions before us are very important.

Answer 4

It does not depend on the volume of lines and databases, but on the types of requests and load. Somewhere infa slipped here that people store 200 million records in MySQL, and there is nothing surprising in this if the data structure is simple and read-only queries, for example.

Answer 5

It seems to me that it depends not only on the volume of the database, but also on the load. If you have a lot of auto-search fields on your site that do a search or other searches with each new letter, then you need to use ES. But not separately. It must be used in conjunction with the base. Let everything work as it works, and search through ES. thus the base will be unloaded.