We* love kubernetes, and all the other orchestration frameworks, like Mesosphere, fleet, docker swarm, etc. We see flocker, our open-source project for managing data volumes, running beside these orchestrators. In fact, we've just released another open-source tool called powerstrip (
https://github.com/ClusterHQ/powerstrip) which makes it possible to write adaptors for the Docker API so that a user can eventually use Flocker with Kubernetes.
Regarding how we are different, we focus only on managing persistent volumes, and as a specialist, will be able to do a lot more than something like kubernetes that is managing the entire application. If I had to guess, what you'd see with kubernetes, mesos, docker, etc is that they all support some basic data movement features, but if you want something more customized, like live migration, or true Disaster Recovery, you'd use Flocker, along side your orchestrator of choice.
*in case it wasn't clear, I work at ClusterHQ