[Gluster-devel] Replicate/AFR Using Broadcast/Multicast?

[Gordan Bobic]

Are there any plans to implement Replicate/AFR using broadcasts (on a 
LAN) or multicast? At the moment, write performance scales inversely 
linearly (O(n)) as additional nodes are added, which gets in the way of 
scalability. It'd be really useful if this could be done using reliable 
multicast (e.g. PGM), as the scaling wouldn't be inverse (it would go to 
O(1)). Are there any plans in this direction?

Gordan