Databases usually are a storage for most applications. Our company also makes a lot of calculations and data manipulations with that data on daily basis.
As soon as we get more and more data, data generation became an issue cause takes too long. And I think it can make sense to separate database to at least two :
-
for storing data with focus on read/write performance;
-
for calculations with focus on data aggregation performance.
Does anybody has similar experience and can tell if this idea is good and what will be design differences for mentioned two points?
Maybe it is worth to look for noSQL solution for calculating data e.g. in-memory databases?
Aucun commentaire:
Enregistrer un commentaire