Can multiple instances access the same dataset?

[Double M]

Hi,

you've got a great project going on - I have a couple of questions which I hope you might be able to answer:

1. Can you set up multiple Cumulus instances (i.e. servers) that all access the same Cassandra cluster? I'd like to load-balance requests to more than one instance in order to scale horizontally.
2. Does cumulus support distributed transactions? Some sort of measure to prevent data loss?
3. Is there a (Java) API to process query results or prepare queries?
4. If no, would the Jena framework be compatible with your SPARQL 1.1 endpoint?

Thanks a lot in advance for your time,

- MM