Click here to Skip to main content
15,885,216 members
Please Sign up or sign in to vote.
3.50/5 (2 votes)
See more:
Hi folks,

I'm searching for a distributed memory library, likely in C++ but C# or Java is also ok, which is not SQL-like, in other words, where the memory is stored on each peer in the network and wehen you change a object inside one peer the memory of all other peers should also change automaticly.
Also it would be nice if the library would allow custom classes as distributed objects.
I seached a while but it seems like there isn't something like that out there.

IDominator
Posted
Comments
Sampath Lokuge 21-Dec-13 9:55am    
Do you need distributed memory caching kind of thing or what ?
Ron Beyer 21-Dec-13 10:19am    
Is this something you've seen before, or something you hope exists?
IDominator 21-Dec-13 13:25pm    
I hope it exists... I have only seen library's that work sql-like

1 solution

I think you can use a distributed hashtable.

https://github.com/sit/dht/wiki[^]

this is a c# implementation

http://sourceforge.net/projects/nchord/[^]
 
Share this answer
 
Comments
IDominator 22-Dec-13 9:55am    
Thank you for your answer. But thats not the solution I search, as the data is distributed across the whole cloud. what I need is that all peers have all data in their own memory and the only messages that they send are the changes the peers make in the memory.
I think this concept is not distributed nor shared memory... I hope something like this exists.
[no name] 22-Dec-13 14:34pm    
Of course it exists.
First, it is easy to modify a distributed hashtable to not split across all nodes but just replicate across them.

But also you might want to try distributed cache. Microsoft AppFabric will do that.(was called Velocity time ago).
IDominator 22-Dec-13 16:01pm    
http://www.codeproject.com/Tips/96984/How-to-Enable-Local-Cache-When-Using-Velocity-Micr
This is almost it. But again it is sql-like, so the memory and the processes that want acess it are seperate and have to comunicate through the network
[no name] 23-Dec-13 11:13am    
I dont understand what you mean by "sql-like".

appFabric is not sql-like (no matter what that is :) ), its a cache mechanism. It might store in sql in the backend, but again, is not SQL-like. You can even use it without sql at all.

If you want "distributed memory" you can use a distributed hashtable and modify the redundancy parameter. that's it :) of course at that time it will be behave more like a 'replicated hashtable' than a distributed one.

IDominator 23-Dec-13 11:22am    
That's it :D I need replicated memory. Do you know a good framework for this, which would also delta-compress it for me?

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)



CodeProject, 20 Bay Street, 11th Floor Toronto, Ontario, Canada M5J 2N8 +1 (416) 849-8900