Why not use couchdb for this use case? Milind /*********************** sent from my android...please pardon occasional typos as I respond @ the speed of thought ************************/
On Aug 18, 2011 9:07 PM, "Nicholas Neuberger" <nneuberg...@gmail.com> wrote: I've been using Cassandra as a database storage device for a service based application. I'm wondering if you can design a multi-tiered cassandra cluster that is used by both clients and servers. I'd like to have the ability to setup the following: *Implement a Core Seed Servers / Nodes internal on network server infrastructure. *Allow client nodes to connect as needed to the Cluster, pickup / synchronize data that they care about, possibly in the same CF, only synch data that they have access to. Client then drops. It's almost like using Cassandra as a synchronization medium, but still store ALL data on the core seed servers / nodes that always run. I'm trying to bridge that gap where you have independent nodes (end clients) that only care about their data, and then reconnect with the mother ship and synchronize their data and drop. With this particular pattern, you could use cassandra as a client / server type of data synchronization. An alternative solution was to send their data via some REST pattern into cassandra, but that seemed like extra work. Does this sound like a good idea, or is there some completely different way of handling data between clients. My goal was to use cassandra (lightweight) on a client application, and on a server / services application and synchronize the data. Ideas? Suggestions? Feature Request? Maybe I fell of the boat into the waters. Nick
