Skip to main content

Network Domains

Waku is a unified and cohesive entity that offers a rich ecosystem with three distinct network interaction domains. These domains serve specialised purposes and contribute to the robust functionality of Waku, forming its foundation.

Discovery domain​

Peer discovery in Waku facilitates locating other nodes within the network. As a modular protocol, Waku incorporates various discovery mechanisms, such as Discv5 and Peer Exchange. These mechanisms allow developers to choose the most suitable option(s) for their specific use cases and user environments, including mobile phones, desktop browsers, servers, and more.

Gossip domain​

GossipSub derives its name from the practice within Pub/Sub networks where peers gossip about the messages they have encountered, thus establishing a message delivery network.

Waku employs gossiping through Relay to distribute messages across the network. Additionally, Waku introduces RLN Relay, an experimental mechanism that combines privacy preservation and economic spam protection.

Request/response domain​

Waku provides a set of protocols to optimise its performance in resource-limited environments like low bandwidth or mostly offline scenarios for multiple purposes.

  • Store enables the retrieval of historical messages.
  • Filter efficiently retrieves a subset of messages to conserve bandwidth.
  • Light Push facilitates message publication for nodes with limited bandwidth and short connection windows.

Overview of protocol interaction​

Here is a diagram illustrating the interaction between different protocols within the Waku Network.

The Pub/Sub topic pubtopic1 serves as a means of routing messages (the network employs a default Pub/Sub topic) and indicates that it is subscribed to messages on that topic for a relay. Node D serves as a Store and is responsible for persisting messages.

  1. Node A creates a WakuMessage msg1 with Content Topic contentTopic1.
  2. Node F requests to get messages filtered by Pub/Sub topic pubtopic1 and Content Topic contentTopic1. Node D subscribes F to this filter and will forward messages that match that filter in the future.
  3. Node A publishes msg1 on pubtopic1. The message is sent from Node A to Node B and then forwarded to Node D.
  4. Node D, upon receiving msg1, stores the message for future retrieval by other nodes and forwards it to Node C.
  5. Node D also pushes msg1 to Node F, informing it about the arrival of a new message.
  6. At a later time, Node E comes online and requests messages matching pubtopic1 and contentTopic1 from Node D. Node D responds with msg1 and potentially other messages that match the query.