1. 3

This works through a gossip protocol for splitting work on node-local queue to other nodes (if local processing capacity is saturated). After nodes run out of work items, a Dijkstra’s token ring is initiated to determine termination. Therefore, this is a self-stabilizing algorithm that can be applied to any general treewalk problem across a distributed system to have near perfect balance on compute workload, regardless of system size.



  2. 1

    To the folks who moderated this as “low quality”, please leave a comment explaining why.

    This work was accepted into an IEEE conference (SC12), so saying that it’s “low quality” is confusing. It’s one of the fastest parallel tree walking algorithms currently known to mankind.


    Perhaps I’m misunderstanding what “low quality” means?

    Edit: It looks like it’s just a mistake. A response from jcs on this complaint can be found at: