[Feature Request] The ability to have discrete blocks of data that can be stored to disk or loaded into memory on command

hackpar / redis

Automatically exported from code.google.com/p/redis

BSD 3-Clause "New" or "Revised" License

0 stars 0 forks source link

I'm going to have a situation where I have very discrete blocks of data per 
user, with up to 10s of millions of rows for power users (but often only a 100k 
rows for casual users). It would be great if I could have a separate Redis 
store for each user, and load their data store from disk into memory when they 
come online, and then back to disk when they go offline. 

With the increased compression of the data when stored to disk, this would seem 
like a very very efficient way to get the most out of each server.  Why have 
any of a user's 10mil rows of data if in memory if  I know he is offline?  
Thoughts?

Original issue reported on code.google.com by xpe...@gmail.com on 23 Jul 2011 at 12:56

Very discrete blocks of data per user sounds like a document oriented store better fits your problem here (like Mongo). Maybe we'll investigate going to disk at a later point in time, but that will likely not concern manually swapping sets of keys to disk and back to memory.

hackpar / redis

[Feature Request] The ability to have discrete blocks of data that can be stored to disk or loaded into memory on command #608