Currently, the KOS fabric operates in a hive-like cluster manner, where data flows propagate without hop limits to all instances in the distributed network.
To support localized propagation scope, we should introduce hop limits for sync/pull/push reactors in order to better control network-wide broadcast storms.
Currently, the KOS fabric operates in a hive-like cluster manner, where data flows propagate without hop limits to all instances in the distributed network.
To support localized propagation scope, we should introduce hop limits for sync/pull/push reactors in order to better control network-wide broadcast storms.