Processing storage of billions of information?

What is the best solution if I need to have a database with a billion + objects, and I need to immediately (or almost immediately) access any of the elements in the database at any time.

This database will be queried at approximately 1000 queries per second. The rows in the database are fairly unrelated and therefore should not be relational.

If you're wondering why, for modeling moving elements.

I was thinking of something like several balanced loads of Cassandra clusters accessed through a load balanced cluster of web servers.

Money is a factor, so the cheaper the better. There are no restrictions on the software or tool, however it must be open.

Just looking for a database solution that would be useful for processing a ridiculous amount of data (generally not necessarily to be relational) by a large number of users.

It is important that it handles redundancy and errors.

Just a high level idea to put me in the right direction would be great.

+3
source share
2 answers

One of the options for consideration is the comparison of three-dimensional coordinates on the curve with a space gap , effectively representing the point as one value. You can then run Cassandra range queries to get points in the area.

I saw how this was implemented in 2D space before, I am sure that this is also possible in 3D.

+1

(X_min <= X_obj <= X_max Y_min <= Y_obj <= Y_max Z_min <= Z_obj <= Z_max), , , Cassandra. , MongoDB, , .

0

Source: https://habr.com/ru/post/1755807/


All Articles