J
J
joger2011-11-12 11:36:02
Database
joger, 2011-11-12 11:36:02

Choosing the right database

At the moment there is a MySQL database consisting of several tables. Each of the tables is a few million rows of mostly text fields. Now the search is on a single indexed field.
Now somewhere between 100-200 thousand requests per day
. The amount of data is constantly increasing, there is a desire to expand the search by several more fields (all - text).
How to approach this issue correctly? Another database, NoSQL, quantum computer?
Thanks in advance for any thoughts or links on this.

Answer the question

In order to leave comments, you need to log in

3 answer(s)
M
mark_ablov, 2011-11-12
@mark_ablov

Sphinx / Solr

R
rPman, 2011-11-12
@rPman

The answer lies in how exactly the search is done in strings? comparison, like substring, full-text index. What is being searched for… a random string or keyword (substrings, predefined and rarely changed, up to a maximum). It is also very important which requests are more and by how much (changes, additions, deletions, reading or searching).
And then, depending on the answers, you can replace the test fields with a group of strongly typed fields (including a complex table structure), you can set up an 'index from your fields' on top of the existing database (by the way, this can allow you to remove the database index, which may in some cases speed up recording), etc.

A
Anatoly, 2011-11-12
@taliban

If you have queries like LIKE "test%", then hit the table for several, let's say a-z, i-t, y-z.

Didn't find what you were looking for?

Ask your question

Ask a Question

731 491 924 answers to any question