HyperDex’s key features – namely its rich API, strong consistency, and fault tolerance – provide strong guarantees to applications that are not matched by other NoSQL systems.
If you find hyperdex interesting, you might like spherical hashing
Looks interesting. Funny though; I ended up looking at the “Contact Us” page and noticed that all of the authors actually seem to be associated with the same university.
I wonder if other universities and the people associated with them tend to more aggressively “brand” projects made by the aforementioned people, or whether it’s more of a culture or a volume thing.
I think HyperDex is unique in this regard. Most research projects have a small web page, and seem dormant soon after the last publication. The end deliverable is a paper, and the code is an artifact of that.
Early on, we realized there was an opportunity to have real users for our project, and put in the effort to make it more than just a hacked-together prototype. This has paid off in spades, as it generates new and interesting research directions that we would not have otherwise thought about had we kept it a small research project with no outside users.
We just released the latest version (1.4.0), which brings performance improvements and a few bug fixes. More information can be found in the release announcement
I have seen a little talk on the mailing list about multi datacenter, have you sketched out a plan for it at all? Riak and Cassandra have a very solid multi datacenter story: the data model lends itself to write-availability and merging of data. But HyperDex takes the strong consistency approach which suggests you’ll have to pay some cost either in latency or availability.