Hi all
I “assume” that if you use an embedded database in your 32 bit application that the db usage would go towards the max amount of memory that a 32 bit app can use?
We have an app that loads some large csv files into memory that we do a lot of searching and grouping on. I’m thinking / hoping that if the csv file were imported into an internal db that the searching and grouping would be faster than having the data in a sorted list and doing a find?
Thanks
Richard
Consider using Postgres, you can silently install it so your users won’t notice or care and then you have some amazing algorithms available to process data faster than you can imagine.
I’ve used Firebird for local datasets over many years (it’s been around forever and I’m still impressed with the latest updates). Never had an issue with large datasets (I think it caches them to disk). And for the embedded option - no local files to deploy.
Probably worth a look ?
Hi Richard
My understanding of an “embedded” database is that the database engine Code is either included in the application exe or linked as a dll or bpl.
Non Embedded (perhaps normal) requires a separate install of the database server probably as a “service”
The database files (data) is I think in most instances the same. The main exception being that a server is better able to manage multiple user access. Embedded Dbs may be single user.
NexusDB can be embedded in your app. You can also have it keep a table just in memory.
The assumption is correct, except if you were to use NexusDB Embedded with our AWE extension. With that combination, your 32 bit process can use as much memory as your 64 bit OS can handle.