Niclas Hedhman wrote: > Does this at all help on the "redundancy" in storage? > > If Entity1 lives on NodeA, then it can't only store itself on NodeA > (assuming NodeA is a single host) as it will get destroyed both in RAM > and disk at the same time. But would it be possible to create small > clusters which shares a redundant store for storage, but where each > Entity only live on a single host? Is that at all smart? And isn't the > "share nothing" philosophy creating new (interesting) challenges for > indexing and searches? (Or does one just accept that Google, Yahoo... > are "almost-infinite scalable")
Redundancy is an interesting and related problem. The way mr Scalability (aka Cameron Purdy of Tangosol/Coherence fame) solves it is by letting each object have one backup node where it is stores synchronously on update. There may also be more nodes backing it up, asynchronously. There's a great presentation by him on InfoQ about these techniques, which might be useful to look at: http://www.infoq.com/presentations/distributed-caching-lessons Note that this is NOT "share nothing", i.e. "duplicate everything", but rather of "as long as the network sort of works, the system will find a way to make it work" approach. Coherence looks at the system more like an organism than a set of separate machines. If it's "up enough", the whole thing will always work. /Rickard _______________________________________________ qi4j-dev mailing list [email protected] http://lists.ops4j.org/mailman/listinfo/qi4j-dev
