All geek questions in one place

The best database solution for managing massive amounts of data

I need to create a traffic database that includes data from different cities (8 cities) 2mb in 10 minutes for each city in 24 hours. The incoming data is the same for all cities. Therefore, my first question is what is better in terms of performance: create one database for all cities with many tables (one table for each city) or create many databases (one database for each city)? My second question is the best database management system for this scenario: MySQL, Postgres, Oracle or others?

+3
source share
2 answers

The amount of data you receive every day is quite a lot (~ 5 GB), but the number of inserted rows is actually quite low. Therefore, you need to develop your physical model in order to simplify the administration of the database warehouse and efficiently execute queries.

Having a separate database in each city makes sense only if you have a server for the database. But you do not need load balancing, as you only need to process eight inserts every ten minutes. On the other hand, this architecture will turn every query that compares one city with another into a distributed query.

, , . . , .

, , (TOWN_ID). , spring Oracle Partitioning.

+3

, . , , , .

- . Anty, , Oracle SQL Server, . , Postgre Mysql , - .

0

Source: https://habr.com/ru/post/1732987/

More articles:


All Articles