experimental distributed hashtable implementations with 2D topology
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

1.3 KiB


status: experimental

Results of our applied research on geo-routing / -hashing in peer to peer systems. This repo provides two DHT implementations with a 2D topology:

  • 2D-CAN: DHT with a 2D grid-based topology. Fully functional, though no geohashing is implemented (unclear how this would be meaningful apart from a translation of geographic coordinates to keyspace coords..). Currently, nodes only know about their direct neighbours.

  • 2D-Kademlia: DHT with XOR-metric based routing. Compared to Kademlia, the address space is twice as deep to accomodate for spatial addressing: Content- & node-addressing works through a [geohash] forming a Z-order curve, allowing to infer spatial relationships between nodes.

Originally, the plan was to compare the routing of both implementation through simulation in The ONE; but we're lacking time for this.


To start the first Node in a DHT:

python app.py

To start a new node and join it to an existing DHT:

python app.py {entry_port}

...where {entry_port} is a port on localhost with another running Node.

Currently nodes only communicate whithin localhost. It should be easy to generalize to full IPs with some small modifications.

Valid commands

  • GET {key}
  • PUT {key} {value}