I need to store 10 years of minute by minute statistics with time index (and make calculation). I don't want a sql db but NoSQL. What would be best for .NET ?
By best I mean:
1°) Performance (for fast reading so as to do calculation in 2) 2°) Easy to do calculation like average since a time origin for example 3°) For .NET
The best for '.NET' is subjective. The only one that is built using .NET is RavenDB, but others like CouchDB and MongoDB also have 'drivers' that lets you query them using a set of .NET libraries so from the client all of them are usable in a .NET application.
Although they all run as well on Windows, I guess it is more related to on which platform you're planning to run them. Check here for the language they are developed in, it may be harder to install and maintain something developed in Erlang in my opinion.
Another point to note is the way you create indexes. In RavenDB you create indexes and map/reduce functions in .NET using Linq. In CoucbDB you use Javascript.
How about using the application cache?
Not really a DB per se but it'll bring performance and it works in a NoSQL Key/Value way. Downside would be storage space but you could build your system to hold both a permanent storage in the DB and a temporary storage in cache for calculations. Other downside if you're working with multiple servers: you need to worry about cache availability over several machines. Some downsides but HUGE performance benefit IMO.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With