Distributed Hash Tables

Description

Rhyzom:

"Distributed Hash Tables store key-value pairs where any node can retrieve the value associated with a given key as a lookup service. Maintaining the key-to-value mappings is distributed among peering nodes.

Such sharded agent-centric model alleviates the burden of having to (similar to lie-detector polygraphs) replicate the same output across all nodes in the network, forming agency around individual actors and the kinds of relationships they enter into instead (each agent storing only what is relevant to him). Agents record and store data in their individual source chains (or hashchains), unfolding parts of them onto the respective app DHTs (private and public space, respectively) and mutually audited (according to the implied validation rules) and countersigned records of transactions with other agents in their source chains.

As mentioned, there is no notion of network-wide consensus (which, by definition, doesn’t scale) bottlenecking the system and strictly limiting the kinds of flows that can enter and the ways in which it can be seen and translated between different contextual interpretations as useful information (and from the reference point of a given agent’s view of the network, or in other words, data is understood as relative to the circumstances at hand, not as an absolute, objective category). DHTs as such constitute a form of sharded validation (what Ethereum is looking to eventually accomplish with PoS+sharding, now known as Ethereum 2.0) where participants store only those shards they happen to validate." (https://rhyzome.github.io/Ceptr-and-Holochain/)