On Tue, Dec 27, 2016 at 2:32 AM, Alexei Scherbakov < alexey.scherbak...@gmail.com> wrote:
> 2016-12-27 10:42 GMT+03:00 Yakov Zhdanov <yzhda...@apache.org>: > > I think the NodeComparator approach will work. User can chose how to sort > > nodes from one rack before nodes from another rack. Same goes for > subnets, > > or data centers. > > >> > > > > Dmitry, can you please explain why you enforce user to write code? This > > does not seem convenient to me at all. If user wants to write code then > he > > can do it for calculating proper arc_id. > > > > Yakov, where is no need to for user to write code. We can provide two > default Comparator implementations: > first based on IP address(default), and second based on node attribute. > User just plugs one of the implementations and adds node attribute to node > config in second case - let it be ARC_ID by default. Completely agree with Alexey here. NodeComparator sounds like a generic approach. We can provide various implementations of comparator with different sorting strategies out of the box.
