Smart L1 cache invalidation

Need to build a mechanism in which L1 invalidation is NOT multicast, but instead is unicast if necessary to specific nodes that may have cached a given entry. This can be detected by maintaining a list of nodes who have requested a key via a remote get, but this would need to be relayed by all data owners.

Benefits would be performance by removing unnecessary invalidation where this is not needed, and by reducing noise in network stacks of most nodes.