riak-mesos

Similar work is progressing well for (off the top of my head) Kafka[1], Riak[2], and MySQL[3] though you'll note they're all quick to point out "don't use this for production yet." I'll be keeping on top of these developments, but for now, managing yourself or leaning on AWS services is definitely the pragmatic decision. [1]: https://github.com/mesos/kafka [2]: https://github.com/basho-labs/riak-mesos [3]: https://github.com/apache/incubator-cotton

https://github.com/basho-labs/riak-mesos is the source link that the site doesn't appear to include. If there are more nodes requested than there are agents available, the scheduler will then start adding more Riak KV nodes to existing agents. It seems to me like this tends to violate the promises you expect from Riak, in that you could end up with data replicated to three nodes that all happen to be on the same underlying hardware. Maybe set safer defaults? I'm excited to see new work coming out of Basho.

Co-author of the code here, if anyone has any questions. The code: https://github.com/basho-labs/riak-mesos