We just released the Hurricane code publicly! The license is Apache 2, so feel free to head over to Github and take it out for a spin!
Hurricane is a high-performance large-scale data analytics system that successfully tames skew in novel ways. Hurricane performs adaptive work partitioning based on load observed by nodes at runtime. Overloaded nodes can spawn clones of their tasks at any point during their execution, with each clone processing a subset of the original data.